WO2016070729A1 - 一种实例化vnf的方法及网元管理设备 - Google Patents

一种实例化vnf的方法及网元管理设备 Download PDF

Info

Publication number
WO2016070729A1
WO2016070729A1 PCT/CN2015/092869 CN2015092869W WO2016070729A1 WO 2016070729 A1 WO2016070729 A1 WO 2016070729A1 CN 2015092869 W CN2015092869 W CN 2015092869W WO 2016070729 A1 WO2016070729 A1 WO 2016070729A1
Authority
WO
WIPO (PCT)
Prior art keywords
vdu
vnf
network element
management device
list
Prior art date
Application number
PCT/CN2015/092869
Other languages
English (en)
French (fr)
Inventor
冯爱娟
侯建星
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2016070729A1 publication Critical patent/WO2016070729A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L12/00—Data switching networks
    • H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
    • H04L12/46—Interconnection of networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/34—Signalling channels for network management communication
    • H04L41/342—Signalling channels for network management communication between virtual entities, e.g. orchestrators, SDN or NFV entities
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/40—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using virtualisation of network functions or resources, e.g. SDN or NFV entities

Definitions

  • the present invention relates to the field of communications, and in particular, to a method for instantiating a VNF and a network element management device.
  • NFV Network Functions Virtualization
  • OSS Operation Support System
  • EMS Network Element Management System
  • VNF Virtual Network Function
  • NFV NFV Infrastructure
  • NFVO NFV Orchestrator
  • VNF Manager VNF Manager
  • VIM Virtual Infrastructure Manager
  • the OSS interacts with the Orchestrator for the operation and maintenance of the network.
  • Data warehouse including related templates and VNF software images.
  • NFV Orchestrator The main function of NFV Orchestrator is to implement network service (NS) lifecycle management (such as deployment/expansion/reduction/downline) in the carrier domain (1 or more data centers), as well as NFVI resource orchestration and policy management. And other functions; NFV Orchestrator describes the template according to the information model of NS (NS Descriptor, NSD) decomposes the need for each VNF and cooperates with VNFM to implement VNF deployment.
  • NS network service
  • NSD Network Descriptor
  • VNFM The main function of VNFM is to realize the life cycle management of VNF, such as deployment/expansion/reduction/downline and other automation capabilities; VNFM decomposes the virtual machine according to VNF's information model description information (VNF Descriptor, VNFD) template and VNF capacity requirements.
  • VNF Descriptor VNF Descriptor
  • VNFC VNF's information model description information
  • VIM The main function of VIM is to manage and monitor the entire infrastructure layer resources (including computing, storage, network resources).
  • Vendor provides input of VNFD parameters in a format understandable by NFV Orchestrator, and related software packages. Operators register the packets in the NFV Orchestrator catalog. .
  • VNFD VNFD in the current NFV specification is very detailed (Table 1 is for reference only), and the details refer to the NVF MAN001GS document.
  • the VNFM is responsible for the life cycle management process of the VNF, and each vendor has different implementations in each process.
  • the VNFD provided by each Vendor must cover the various processes, and Can be standardized, can not reflect the differences of the various vendors.
  • the VNFM receives a Lifecycle-managed operation (the trigger can be NFVO, VNFM itself, EMS, etc., parsing VNFD, and then completing the corresponding actions step by step according to the corresponding parameters or procedures in VNFD and the instructions of the script, making the VNF lifecycle management process very mechanized and lacking flexibility.
  • the embodiment of the invention provides a method for instantiating a VNF and a network element management device, which can simplify the process of VNF instantiating VNF, and can support the NE management device of each vendor to perform VNF instantiation differential management.
  • a first aspect of the present invention provides a method of instantiating a virtualized network function VNF, which may include:
  • the network element management device receives the instantiated VNF request message sent by the VNF manager VNFM.
  • the network element management device instantiates the VNF according to the determined resource.
  • the determining, by the network element management device, the resources required to perform the instantiation of the VNF, according to the instantiated VNF request message and the resource configuration information stored in the network element management device specifically includes:
  • the network element management device Determining, by the network element management device, the virtual deployment unit VDU list required to instantiate the VNF according to the instantiated VNF request message and the resource configuration information stored in the network element management device, and required for each VDU in the VDU list Virtual machine VM resource;
  • the method for the network element management device to instantiate the VNF according to the determined resource specifically includes:
  • the network element management device generates a command list for deploying the VDU according to the determined VDU list
  • the network element management device requests the virtual machine VM resource required by each VDU in the command list to the virtualized infrastructure management system VIM in the process of executing the deployment command of the VDU in the command list, and according to the VIM Assessing the virtual machine VM resources required by each VDU in the command list, loading software for each VDU in the VDU list and performing basic data configuration;
  • the network element management device After the deployment command of all the VDUs in the VDU list is executed, the network element management device notifies the VNFM that the instantiated VNF is completed.
  • the network element management device In the process of executing the deployment command of the VDU in the command list, the network element management device requests the virtual machine VM resource required by each VDU in the command list from the VIM, and the command is performed according to the VIM.
  • the result of the allocation of the virtual machine VM resources required by each VDU in the list, the software loading and the basic data configuration for each VDU in the VDU list specifically includes:
  • the network element management device executes the deployment command of each VDU in the command list one by one.
  • the virtual machine VM resource required for each VDU is allocated to the VIM request.
  • loading software for each VDU according to the resource allocated by the VIM for each VDU, and allocating a required virtual machine VM resource allocation result to each VDU according to the VIM, for each of the The VDU loads the software and performs basic data configuration.
  • a second aspect of the present invention provides a network element management device, which may include:
  • a receiving module configured to receive an instantiated VNF request message sent by the VNF manager VNFM;
  • a determining module configured to determine, according to the instantiated VNF request message received by the receiving module and the resource configuration information stored in the network element management device, a resource required to execute the instantiated VNF;
  • an instantiation module configured to instantiate the VNF according to the resource determined by the determining module.
  • the network element management device receives the instantiated VNF request message sent by the VNF manager VNFM, and determines to perform according to the instantiated VNF request message and the resource configuration information stored in the network element management device.
  • the resource required for instantiating the VNF; and the instantiation of the VNF according to the determined resource, the network element management device and the VNF belong to the same vendor, and the network element management device between different vendors does not need to be standardized. And the unified embodiment. Therefore, the embodiment of the present invention can support different vendors in performing the instantiation of the VNF. For example, the format of the resource configuration information stored in different vendors may be different, and the different vendors determine the basis.
  • the specific process of instantiating a VNF can also be different.
  • the embodiment of the present invention transfers the lifecycle management work of the VNFM to the network element management device, thereby eliminating the need to set a complete parameter or script that can cover each VNF lifecycle management process in the VNFD.
  • the network element management device does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to perform the lifecycle management of the instantiated VNF.
  • the action should be taken to avoid over-mechanization and lack of flexibility in the lifecycle management process of instantiating VNF.
  • Figure 1 shows the reference framework of Network Functions Virtualization (NFV) defined by NFV ISG;
  • NFV Network Functions Virtualization
  • FIG. 2 is a schematic flowchart of a method for instantiating a virtualized network function VNF according to an embodiment of the present invention
  • step S111 in the method flow shown in FIG. 2;
  • step S112 in the method flow shown in FIG. 2;
  • FIG. 5 is a schematic flowchart diagram of an embodiment of a method for expanding a VNF instance according to an embodiment of the present disclosure
  • FIG. 6 is a schematic flow chart of an embodiment of step S211 in the method flow shown in FIG. 5;
  • FIG. 7 is a schematic flowchart diagram of another embodiment of a method for expanding a VNF instance according to an embodiment of the present disclosure
  • step S221 in the method flow shown in FIG. 7;
  • step S212 in the method flow shown in FIG. 5 or FIG. 7;
  • FIG. 10 is a schematic flowchart diagram of an embodiment of a method for shrinking a VNF instance according to an embodiment of the present invention.
  • step S311 is a schematic flow chart of an embodiment of step S311 in the method flow shown in FIG. 10;
  • FIG. 12 is a schematic flowchart diagram of another embodiment of a method for shrinking a VNF instance according to an embodiment of the present invention.
  • FIG. 13 is a schematic flowchart of an embodiment of step S321 in the method flow shown in FIG. 12;
  • step S312 is a schematic flow chart of an embodiment of step S312 in the method flow shown in FIG. 10 or FIG. 12;
  • FIG. 15 is a schematic flowchart diagram of a method for terminating a VNF instance according to an embodiment of the present invention.
  • step S411 in the method flow shown in FIG. 15;
  • step S412 in the method flow shown in FIG. 15;
  • FIG. 18 is a schematic structural diagram of an embodiment of a network element management device according to an embodiment of the present invention.
  • FIG. 19 is a schematic structural diagram of an embodiment of the determining module 11 of FIG. 18;
  • FIG. 20 is a schematic structural diagram of an embodiment of the instantiation module 12 of FIG. 18;
  • FIG. 21 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention.
  • 22 is a schematic structural diagram of an embodiment of the determining module 21 of FIG. 21;
  • FIG. 23 is a schematic structural diagram of an embodiment of the expansion module 22 of FIG. 21;
  • FIG. 24 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention.
  • 25 is a schematic structural diagram of an embodiment of the determining module 24 of FIG. 24;
  • FIG. 26 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention.
  • FIG. 27 is a schematic structural diagram of an embodiment of the determining module 31 of FIG. 26;
  • FIG. 28 is a schematic structural diagram of an embodiment of a reduction module 32 of FIG. 26;
  • FIG. 29 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention.
  • FIG. 30 is a schematic structural diagram of an embodiment of the determining module 34 of FIG. 29; FIG.
  • FIG. 31 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention.
  • FIG. 32 is a schematic structural diagram of an embodiment of the determining module 41 of FIG. 31;
  • 33 is a schematic structural diagram of an embodiment of the termination module 42 of FIG. 31;
  • FIG. 34 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention.
  • FIG. 35 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention.
  • FIG. 36 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present disclosure.
  • FIG. 37 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention.
  • FIG. 39 is a schematic flow chart showing the process of terminating a VNF instance according to the present invention.
  • FIG. 2 is a schematic flowchart diagram of a method for instantiating a virtualized network function VNF according to an embodiment of the present invention. As shown in FIG. 2, it may include:
  • Step S110 The network element management device receives the instantiated VNF request message sent by the VNF manager VNFM.
  • the network element management device and the VNF in the embodiment of the present invention are provided by the same vendor, and may include, but are not limited to, an Element Management System (EMS) or VNF Local Management Unit (LMU).
  • EMS Element Management System
  • LMU VNF Local Management Unit
  • the network element management device may receive an instantiated VNF request message from the VNFM through a Ve-Vnfm interface with the VNFM.
  • Step S111 the network element management device determines, according to the instantiated VNF request message and the resource configuration information stored in the network element management device, resources required to perform the instantiation of the VNF.
  • the network element management device determines, according to the instantiated VNF request message and the resource configuration information stored in the network element management device, a virtual deployment unit VDU list required to instantiate the VNF, and the VDU list.
  • the virtual machine VM resource may include: a VM VM specification required by each VDU, a requirement of each VM to a link, and the like.
  • the resources required to perform the instantiation of the VNF determined by the embodiment of the present invention may further include: software corresponding to each VDU and affinity and isolation of each VM.
  • step S111 may include the following steps:
  • Step S1110 The network element management device determines, according to the request for instantiating the VNF, a type, a version number, and a capacity specification of the VNF to be deployed.
  • the network element management device may describe the template VNFD index information and the capacity specification according to the information model of the VNF carried in the instantiated VNF request, and query the corresponding VNFD for the type of the VNF to be deployed. , version number and capacity specifications. For example, if the VNFD index carried in the request for instantiating the VNF is the index xx of the VNFD of Table 1, and the capacity specification is “Deployment Specification 1” in Table 1, the network element management device may be configured according to the step S1110.
  • VNF Index xx finds the VNFD described in Table 1, and in Table 1, can query the type of VNF to be deployed (corresponding to "VNF id” in Table 1), version number (corresponding to "VNF version” in Table 1), and Capacity specification (corresponding to "Deployment Specification 1" in Table 1).
  • the carrying of the capacity specification may only carry the capacity specification number (corresponding to the number 1 of the deployment specification 1 in Table 1), and may also carry the capacity size of the capacity specification (corresponding to the capacity of 500 kcaps in Table 1).
  • the network element management device may determine the type, version number, and capacity specification of the VNF to be deployed according to the type, version number, and capacity specification of the VNF carried by the instantiated VNF request.
  • the type of the VNF carried by the request for instantiating the VNF is CSCF
  • the version number is 1
  • the capacity specification is deployment specification 1 (corresponding to the list item of the deployment specification 1).
  • the carrying capacity of the grid can only carry the capacity specification number (for example, the number 1 of the deployment specification 1 in Table 1), or the capacity size of the capacity specification (for example, the capacity of the deployment specification 1 in Table 1 is 500 kcaps).
  • the type of the VNF to be deployed is CSCF
  • the version number is 1
  • the capacity specification is 500 kcaps.
  • the type, version number, and capacity specification of the VNF carried by the instantiated VNF request are registered in the VNFD or registered in an information model description template NSD of the network service.
  • step S1111 the network element management device determines the VDU list required for the VNF to be deployed according to the ratio of the capacity specification in the resource configuration information to the virtual deployment unit VDU.
  • the resource configuration information stored by the network element management device includes a ratio of the capacity specification to the virtual deployment unit VDU (for example, "1*VDU_A+1*VDU_B+1" in the VDU list item in Table 1. *VDU_C"), therefore, in step S1111, the network element management device can determine the capacity specification or capacity of the number 1 according to the correspondence between the stored capacity 500kcaps and the VDU list 1*VDU_A+1*VDU_B+1*VDU_C.
  • the VDU list corresponding to the capacity specification of 500kcaps is 1*VDU_A+1*VDU_B+1*VDU_C.
  • the embodiment of the present invention can simplify the specific content of the capacity specification in the VNFD.
  • the capacity specification column in the VNFD can no longer require the information of the VDU list.
  • VDU_A, VDU_B and VDU_C represent different types of VDUs.
  • step S1112 the network element management device determines software corresponding to each VDU in the VDU list according to the correspondence between the VDU and the software in the resource configuration information.
  • Step S1113 The network element management device determines, according to the correspondence between the VDU and the virtual machine (VM) in the resource configuration information, the VM specifications required by each VDU.
  • the VM specifications required for different types of VDUs may be different. Examples are as follows:
  • the VM specifications required for one VDU_A include: 1 VM_A, corresponding to 4 core vCPU + 8G memory + 2G bandwidth;
  • the VM specifications required for one VDU_B include: 1 VM_B, corresponding to 4 core vCPU + 16G memory + 2G bandwidth;
  • the VM specifications required for one VDU_C include: 1 VM_C, corresponding to 2 core vCPU + 8G memory + 2G bandwidth.
  • Step S1114 The network element management device determines, according to the network interworking relationship in the resource configuration information and the bandwidth requirement of each link, the requirements of each VM for the link.
  • the interworking relationship and bandwidth requirements of the network elements in the resource configuration information stored in the NE management device can be as follows:
  • VDU_A and VDU_B are interworking (that is, VM_A and VM_B are interworking), and the bandwidth needs to be 2G.
  • VDU_B is interworking with VDU_C (that is, VM_B and VM_C are interworking), and the bandwidth needs to be 2G.
  • the link requirement determined in step S1114 may be a virtual link between VM_A and VM_B, and a virtual link needs to be established between VM_B and VM_C.
  • step S111 may further include step S1115, and the network element management device determines the affinity and isolation of the VM according to the VDU isolation and affinity in the resource configuration information.
  • the VDU isolation and affinity in the resource configuration information stored by the network element management device can be as follows:
  • VDU_A and VDU_C are deployed on the same physical server
  • VDU The same type of VDU is deployed on different physical servers.
  • the affinity and isolation of each VM can be determined according to the VDU isolation and affinity stored by the network element management device.
  • VM_A and VM_C are deployed on the same physical server, and multiple VM_As are deployed in different physical environments.
  • the types of VM_1 and VM_2 are VM_A, but when deployed, VM_1 and VM_2 need to be deployed on different physical servers.
  • the embodiment of the present invention may create an instance identifier for the VNF instance that is requested to be established by the instantiated VNF request message in the network element management device, and establish the VNF instance. Correspondence between the version number, type, and capacity specifications and the instance ID.
  • Step S112 the network element management device instantiates the VNF according to the determined resource.
  • step S112 may specifically include the following steps:
  • Step S1120 The network element management device generates a command list for deploying a VDU according to the determined VDU list.
  • each type of configuration command is stored in the network element management device of the embodiment of the present invention. For example, deploy the command ADD VDU; delete the command RMV VDU. Therefore, in step S1120, the network element management device may separately generate a deployment command for each VDU in the determined VDU list, and the deployment command set generated by each VDU in the entire VDU list forms a VDU deployment command list.
  • the embodiment of the present invention may generate the following VDU deployment command list (the corresponding VDU list is 1*VDU_A+1*VDU_B+1*VDU_C):
  • VDU_1 (representing the number of the deployed VDU), VDU_A (representing the deployed VDU type);
  • VDU_2 (representing the number of the deployed VDU), VDU_B (representing the deployed VDU type);
  • VDU_3 (representing the number of the deployed VDU), VDU_C (representing the deployed VDU type);
  • step S1121 the network element management device requests, from the VIM, the virtual machine VM resources required by each VDU in the command list in the process of executing the deployment command of the VDU in the command list, and according to the VIM As a result of the allocation of virtual machine VM resources required by each VDU in the command list, software is loaded for each VDU in the VDU list and basic data is configured.
  • the VM resource and the link resource determined according to the processes of steps S1110-S1115 may be used to allocate corresponding resources to the VIM, and the software is loaded and basic data is configured for each VDU in the VDU list.
  • the performing the basic data configuration means that the software loaded for each VDU, the allocation result of the virtual machine VM resources allocated by each VDU, and the type of each VDU are recorded in the network element management device, and Each VDU is associated with an identifier of a VNF instance including the VDUs as a subsequent available resource configuration information, so that the network element management device manages the VNFs in the VNF instance and the instance based on the basic data configuration. It can be seen that the embodiment of the present invention automatically completes the basic data configuration in the process of performing the VNF instantiation, and solves the technical problem that the prior art needs to manually perform basic data configuration in the network element management device.
  • the network element management device may execute a deployment command of each VDU in the command list one by one, and allocate each of the VDUs to the VIM request during a deployment command of each VDU.
  • Required virtual machine VM resources and according to the VIM for each of the VDUs The allocated resources load software for the one VDU and allocate software for each VDU according to the VIM to allocate the required virtual machine VM resource allocation result according to the VIM, and perform basic data configuration for each VDU.
  • the embodiment of the present invention may request the VM resource and the link resource required by all the VDUs in the VDU list to the VIM at one time, and load the software for each VDU once after the request is completed. And perform basic data configuration.
  • Step S1122 After the deployment command of all VDUs in the VDU list is executed, the network element management device notifies the VNFM that the instantiated VNF is completed.
  • the embodiment of the present invention may further include the following steps:
  • the network element management device initiates an operation authorization request for instantiating the VNF to the Network Function Virtualization Orchestrator (NFVO) via a VNFM (VNFM), and the operation authorization request is performed by the VNFM
  • NFVO receives a response to the operation authorization of the instantiated VNF
  • the response of the operation authorization of the instantiated VNF is that the NFVO determines that there is a resource required to execute the instantiated VNF and requests the VIM to reserve a corresponding resource and then send of;
  • the network element management device in the step S1121 requests the VIM to allocate the VM resources and the link resources required for each VDU in the VDU list, and specifically includes:
  • the network element management device forwards, by the VNFM, the request for allocating VM resources required for each VDU in the VDU list to the VIM.
  • the network element management device in the step S1121 requests the VIM to allocate the VM resources required for each VDU in the VDU list, and specifically includes:
  • the network element management device forwards, by the VNFM and the NFVO, a request for allocating VM resources required for each VDU in the VDU list to the VIM.
  • the difference between this embodiment and the previous embodiment is that the NFVO directly initiates a resource allocation request to the VIM, and the previous embodiment further includes an operation authorization request and a response to determine that there are related resources in the network before initiating the resource allocation request.
  • the network element management device receives the real message sent by the VNF manager VNFM. Instantiating a VNF request message; and determining, according to the instantiated VNF request message and resource configuration information stored in the network element management device, a resource required to execute the instantiated VNF; and instantiating the VNF according to the determined resource .
  • the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the network element management device does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to perform the lifecycle management of the VNF, and the corresponding actions are completed step by step, thereby avoiding the over-mechanization of the lifecycle management process of the instantiated VNF. Lack of flexibility.
  • the network element management device and the VNF usually belong to the same provider. Therefore, the embodiment of the present invention can support the difference in the performance of the VNF generated by each vendor, for example, the resource configuration information stored in the network element management device of different vendors.
  • the format may be different, and the specific processes of the network element management device of different vendors to instantiate the VNF according to the determined resources may also be different.
  • FIG. 5 is a schematic flowchart diagram of a method for expanding a VNF instance according to an embodiment of the present invention.
  • the VNF instance expansion in the embodiment of the present invention may be performed independently of the process of instantiating the VNF, or after the foregoing VNF instantiation process of the present invention. As described in FIG. 5, it may include:
  • Step S210 The network element management device acquires performance parameters reported by the VNF.
  • the performance parameter reported by the VNF obtained by the embodiment of the present invention may include a CPU usage rate, or any other performance parameter that can be reported by the VNF.
  • Step S211 the network element management device determines, according to the reported performance parameter and the resource configuration information stored in the network element management device, a VDU list that needs to be deployed to perform VNF instance expansion, and each VDU in the VDU list. Virtual machine VM resources required.
  • the embodiment of the present invention may maintain a VNF instance expansion and contraction policy information in the resource configuration information stored in the network element management device, or maintain a VNF instance expansion strategy information in the resource configuration information.
  • a VNF instance shrinkage policy information where the VNF instance expansion policy may include: when the performance parameter of the last report is lower than the second threshold, and the performance parameter of the current report is greater than or equal to the second threshold, executing the VNF instance
  • the capacity expansion is performed by deploying one or more specified types of VDUs based on the existing VNF instance capacity specifications.
  • the VNF instance shrinking policy may include: when the last reported performance parameter is greater than the first threshold, the present report is performed. Performance parameter is less than the first threshold and When buffering the difference of the parameters, performing VNF instance shrinking, terminating one or more specified types of VDUs in the VNF instance.
  • the VNF instance expansion and contraction policy information in the resource configuration information of the embodiment of the present invention may specifically include (the first two are VNF instance expansion strategy, the last two are VNF instance shrinkage strategies):
  • the current CPU usage of the current report is less than 60% (corresponding to the second threshold).
  • a VDU of type B is expanded.
  • the CPU usage rate of the current report is higher than 60%.
  • the CPU usage of the current report is ⁇ (60%-5%)
  • the VDU of type A is reduced.
  • the CPU usage rate of the current report is higher than 20% (corresponding to the above first threshold).
  • ⁇ (20%-5% (corresponding to the above buffer parameter)) a type is constricted.
  • step S211 when the occupancy rate of the CPU reported by the VNF meets the first policy or the second policy of the foregoing four policies, the VDU list that needs to be deployed for performing the VNF instance expansion may be determined, for example, when the above is satisfied. In the case of the policy 1, it is determined that a VDU of type A needs to be deployed on the basis of the original instance of the VNF.
  • the resource configuration information stored in the network element management device may further include a ratio of the capacity specification to the virtual deployment unit VDU, a correspondence between the VDU and the software, and a corresponding relationship between the VDU and the virtual machine. Or information about network interworking relationships and bandwidth requirements of each link. Therefore, in the embodiment of the present invention, after determining the VDU list to be deployed, the virtual machine VM resources required for each VDU in the VDU list may be further determined according to the resource configuration information stored in the network element management device.
  • the virtual machine VM resource may include: a VM VM specification required by each VDU, a requirement of each VM to a link, and the like.
  • the resources determined by the embodiments of the present invention may further include: software corresponding to each VDU and affinity and isolation of each VM.
  • step S211 of the embodiment of the present invention may specifically include:
  • Step S2110 The network element management device determines, according to the reported performance parameter and the VNF instance expansion policy information in the resource configuration information stored in the network element management device, the VDU list that needs to be deployed to perform the VNF instance expansion. Specifically, how to determine the VDU list refers to the previous description, and details are not described herein.
  • Step S2111 The network element management device determines software corresponding to each VDU in the VDU list that needs to be deployed in the VNF instance expansion according to the corresponding relationship between the VDU and the software in the resource configuration information.
  • the network element management device determines, according to the corresponding relationship between the VDU and the virtual machine (VM) in the resource configuration information, the VMs required for each VDU in the VDU list to be deployed in the VNF instance expansion.
  • VM virtual machine
  • the VM specifications required for different types of VDUs may be different. Examples are as follows:
  • the VM specifications required for one VDU_A include: 1 VM_A, corresponding to 4 core vCPU + 8G memory + 2G bandwidth;
  • the VM specifications required for one VDU_B include: 1 VM_B, corresponding to 4 core vCPU + 16G memory + 2G bandwidth;
  • the VM specifications required for one VDU_C include: 1 VM_C, corresponding to 2 core vCPU + 8G memory + 2G bandwidth.
  • Step S2113 The network element management device needs the link of the VM according to the network interworking relationship in the resource configuration information and the bandwidth requirement of each link.
  • the interworking relationship and bandwidth requirements of the network elements in the resource configuration information stored in the NE management device can be as follows:
  • VDU_A and VDU_B are interworking (that is, VM_A and VM_B are interworking), and the bandwidth needs to be 2G.
  • VDU_B is interworking with VDU_C (that is, VM_B and VM_C are interworking), and the bandwidth needs to be 2G.
  • step S211 may further include step S2114, where the network element management device determines the affinity and isolation of the VM according to the VDU isolation and affinity in the resource configuration information.
  • the VDU isolation and affinity in the resource configuration information stored by the network element management device can be as follows:
  • VDU_A and VDU_C are deployed on the same physical server
  • VDU The same type of VDU is deployed on different physical servers.
  • the affinity and isolation of each VM can be determined according to the VDU isolation and affinity stored by the network element management device.
  • VM_A and VM_C are deployed on the same physical server, and multiple VM_A are deployed in different physical environments.
  • the types of VM_1 and VM_2 are VM_A, but when deployed, VM_1 and VM_2 need to be deployed on different physical servers.
  • steps S210-S211 of the embodiment of the present invention may be replaced by the following process:
  • Step S220 The network element management device receives the VNF instance expansion request from the VMFM.
  • step S221 the network element management device determines, according to the received VNF instance expansion request and the resource configuration information stored by the network element management device, a VDU list to be deployed and a VDU in the VDU list.
  • the required virtual machine VM resources are included in step S221.
  • step S221 may include:
  • Step S2210 The network element management device determines, according to the VNF instance identifier carried in the VNF instance expansion request and the capacity specification after the capacity expansion, the VDU list to be deployed.
  • Step S2211 The network element management device determines, according to the corresponding relationship between the VDU and the software in the resource configuration information, software corresponding to each VDU in the VDU list to be deployed in the VNF instance expansion.
  • step S2212 the network element management device determines, according to the correspondence between the VDU and the virtual machine (VM) in the resource configuration information, the VMs required for each VDU in the VDU list to be deployed in the VNF instance expansion. specification;
  • Step S2213 The network element management device needs the link of the VM according to the network interworking relationship in the resource configuration information and the bandwidth requirement of each link.
  • step S2211 and S2212-S2213 there may be no sequential execution order between step S2211 and S2212-S2213.
  • step S2114 may be further included, and the network element management device determines the affinity and isolation of the VM according to the VDU isolation and affinity in the resource configuration information.
  • Steps S2211 to S2214 are the same as steps S2111 to S2114, respectively, and therefore, the specific implementation of step S2211-step S2214 will not be described again.
  • Step S212 The network element management device performs VNF instance expansion according to the determined VDU list and virtual machine VM resources required by each VDU in the VDU list.
  • step S212 may specifically include the following steps:
  • Step S2120 The network element management device generates a command list for deploying a VDU according to the determined VDU list. It is assumed that the VDU list determined in step S211 or S221 includes one VDU of type A (simplified as: 1*VDU_A).
  • the network element management device of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a deletion command RMV VDU, and the like. Therefore, in step S2120, the network element management device may generate a deployment command for each VDU in the determined VDU list, and the deployment command set generated by each VDU in the entire VDU list forms a VDU deployment command list.
  • the determined VDU list is: 1*VDU_A
  • the embodiment of the present invention may generate the following VDU deployment command list:
  • VDU_1 (representing the number of the deployed VDU), VDU_A (representing the deployed VDU type);
  • step S2121 the network element management device requests, from the VIM, the virtual machine VM resources required by each VDU in the command list in the process of executing the deployment command of the VDU in the command list, and according to the VIM As a result of the allocation of virtual machine VM resources required by each VDU in the command list, software is loaded for each VDU in the VDU list and basic data is configured.
  • the performing the basic data configuration means that the software loaded for each VDU, the allocation result of the virtual machine VM resources allocated by each VDU, and the type of each VDU are recorded in the network element management device, and Each VDU is associated with an identifier of a VNF instance including the VDUs as a subsequent available resource configuration information, so that the network element management device manages the VNFs in the VNF instance and the instance based on the basic data configuration.
  • the basic data configuration is automatically completed in the process of performing the VNF instance expansion in the embodiment of the present invention, which solves the technical problem that the prior art needs to manually perform basic data configuration in the network element management device.
  • the network element management device may execute the command column one by one. a deployment command of each VDU in the table, in the process of executing the deployment command of each VDU, requesting, by the VIM, the virtual machine VM resources required for each VDU, and according to the VIM, each of the VIMs
  • the resources allocated by the VDU load software for the one VDU and allocate the required virtual machine VM resource allocation results to each VDU according to the VIM, load software for each VDU and perform basic data configuration.
  • the embodiment of the present invention may request the VM resource and the link resource required by all the VDUs in the VDU list to the VIM at one time, and load the software for each VDU once after the request is completed. And perform basic data configuration.
  • Step S2122 After the deployment command of all the VDUs in the VDU list is executed, the network element management device notifies the VNFM that the VNF instance expansion is completed.
  • the embodiment of the present invention may further include the following steps:
  • the network element management device initiates an operation authorization request for instantiating the VNF to the Network Function Virtualization Orchestrator (NFVO) via a VNFM (VNFM), and the operation authorization request is performed by the VNFM
  • NFVO receives a response to the operation authorization of the instantiated VNF
  • the response of the operation authorization of the instantiated VNF is that the NFVO determines that there is a resource required to execute the instantiated VNF and requests the VIM to reserve a corresponding resource and then send of;
  • the network element management device in the step S2121 requests the VIM to allocate the VM resources required for each VDU in the VDU list, and specifically includes:
  • the network element management device forwards, by the VNFM, the request for allocating VM resources required for each VDU in the VDU list to the VIM.
  • the network element management device in the step S2121 requests the VIM to allocate the VM resources required for each VDU in the VDU list, and specifically includes:
  • the network element management device forwards, by the VNFM and the NFVO, a request for allocating VM resources required for each VDU in the VDU list to the VIM.
  • the difference between this embodiment and the previous embodiment is that the NFVO directly initiates a resource allocation request to the VIM, while the previous embodiment initiates the resource allocation.
  • the request authorization request and the response also include determining the relevant resources in the network before the request.
  • the network element management device obtains the performance parameter reported by the VNF or the received VNF instance expansion request; and according to the reported performance parameter or the received VNF instance expansion request and the storage in the network element management device
  • the resource configuration information is used to determine the VDU list to be deployed in the VNF instance expansion and the VM VM resources required for each VDU in the VDU list; and perform VNF instance expansion according to the determined resources.
  • the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the lifecycle management of the VNF instance expansion does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to complete the corresponding actions step by step. This avoids the over-mechanization of the lifecycle management process of the VNF instance expansion. Lack of flexibility.
  • the network element management device and the VNF are generally the same vendor. Therefore, the embodiment of the present invention can support the difference in the VNF instance expansion of each vendor, for example, the resource configuration information stored in the network element management device of different vendors.
  • the format of the VNF instance expansion may be different according to the determined resources.
  • FIG. 10 is a schematic flowchart diagram of a method for shrinking a VNF instance according to an embodiment of the present invention.
  • the VNF instance shrinkage in the embodiment of the present invention may be performed independently of the above-described process of instantiating the VNF, or after the aforementioned VNF instantiation process of the present invention. As described in FIG. 10, it may include:
  • Step S310 the network element management device acquires performance parameters reported by the VNF
  • the performance parameter reported by the VNF obtained by the embodiment of the present invention may include a CPU usage rate, or any other performance parameter that can be reported by the VNF.
  • Step S311 the network element management device determines, according to the reported performance parameter and the VNF instance shrinkage policy in the resource configuration information, a VDU list to be terminated by the VNF instance shrinkage and corresponding to each VDU in the VDU list.
  • the VM resource to be released includes one or more VDUs to be terminated in the VDU list to be terminated, and the VNF instance shrinking policy in the resource configuration information includes: when the performance parameter of the last report is greater than the first threshold When the performance parameter reported in this time is smaller than the difference between the first threshold and a buffer parameter, the VNF instance is compressed to terminate one or more specified types of VDUs.
  • the embodiment of the present invention may be configured in the resource configuration information stored in the network element management device. Maintaining a VNF instance expansion and contraction policy information, or maintaining a VNF instance expansion policy information and a VNF instance reduction policy information in the resource configuration information, where the VNF instance expansion policy may include: If the performance parameter of the report is lower than the second threshold, the performance of the VNF instance is expanded, that is, one or more specified types are deployed based on the existing VNF instance capacity specifications.
  • the VNF instance shrinkage policy may include: when the performance parameter of the last report is greater than the first threshold, and the performance parameter of the current report is smaller than the difference between the first threshold and a buffer parameter, performing VNF instance shrinkage, Terminate one or more VDUs of the specified type in the VNF instance.
  • the VNF instance expansion and contraction policy information in the resource configuration information of the embodiment of the present invention may specifically include (the first two are VNF instances), and the performance parameter of the VNF is the CPU usage. Expansion strategy, the last two are VNF instance shrinkage strategies):
  • the current CPU usage of the current report is less than 60% (corresponding to the second threshold).
  • a VDU of type B is expanded.
  • the CPU usage rate of the current report is higher than 60%.
  • the CPU usage of the current report is ⁇ (60%-5%)
  • the VDU of type A is reduced.
  • the CPU usage rate of the current report is higher than 20% (corresponding to the above first threshold).
  • ⁇ (20%-5% (corresponding to the above buffer parameter)) a type is constricted.
  • step S311 when the occupancy rate of the CPU reported by the VNF meets the third policy or the fourth policy of the foregoing four policies, the VDU list required to perform the VNF instance shrinkage may be determined, for example, when the content is satisfied. In the foregoing strategy 3, it is determined that a VDU of type A needs to be terminated on the basis of the original instance of the VNF.
  • the resource configuration information stored in the network element management device may further include a ratio of the capacity specification to the virtual deployment unit VDU, a correspondence between the VDU and the software, and a VDU. Information such as the correspondence with the virtual machine, the network interworking relationship, and the bandwidth requirements of each link. Therefore, in the embodiment of the present invention, after determining the VDU list to be terminated, the virtual machine VM to be released by each VDU in the VDU list may be further determined according to the resource configuration information stored in the network element management device. Resources and software that each VDU needs to delete.
  • the virtual machine VM resource may include: a VM VM specification required by each VDU, a requirement of each VM to a link, and the like. Further, as an additional embodiment, the embodiment of the present invention may further determine, in step S311, a VDU having an affinity and isolation relationship with the determined VDU to be terminated.
  • step S311 of the embodiment of the present invention may specifically include:
  • Step S3110 The network element management device determines, according to the reported performance parameter and the VNF instance shrinkage policy information in the resource configuration information stored in the network element management device, a list of VDUs that need to be terminated to perform VNF instance shrinkage. . Specifically, how to determine the previous description is not described here.
  • Step S3111 The network element management device determines, according to the basic data configuration information of each VDU in the resource configuration information, software that needs to be deleted for each VDU in the VDU list to be terminated, a VM specification to be released, and a release that needs to be released.
  • the link associated with the VM is the basic data configuration information of each VDU in the resource configuration information.
  • VM specifications corresponding to different types of VDUs may be different. Examples are as follows:
  • the VM specifications required for one VDU_A include: 1 VM_A, corresponding to 4 core vCPU + 8G memory + 2G bandwidth;
  • the VM specifications required for one VDU_B include: 1 VM_B, corresponding to 4 core vCPU + 16G memory + 2G bandwidth;
  • the VM specifications required for one VDU_C include: 1 VM_C, corresponding to 2 core vCPU + 8G memory + 2G bandwidth.
  • VDU_A when it is determined that one VDU_A needs to be terminated, it may be determined that one VM_A needs to be deleted, and deleting the VM_A is specifically required to release 4 cores vCPU+8G memory+2G bandwidth.
  • VDU_A and VDU_B are interworking (that is, VM_A and VM_B are interworking), and the bandwidth needs to be 2G.
  • VDU_B is interworking with VDU_C (that is, VM_B and VM_C are interworking), and the bandwidth needs to be 2G.
  • step S3111 the embodiment of the present invention can determine that the link between VM_A and VM_B needs to be released according to the foregoing interworking relationship, and release the link between VM_B and VM_C.
  • the basic data of the VDU configured in the network element management unit may be deleted, for example, Deleting the relationship between the VDU and the VNF instance, deleting the corresponding relationship between the VDU and the software, deleting the correspondence between the VDU and the VM specification, and deleting the link information of the VM of the VDU. .
  • step S311 may further include step S3112, where the network element management device determines, according to the VM isolation and affinity in the resource configuration information, that the VDU in the VDU list to be terminated has The VDU of affinity and isolation.
  • the VM isolation and affinity in the resource configuration information stored by the network element management device may be as follows:
  • VM_A and VM_C are deployed on the same physical server
  • the same type of VM is deployed on different physical servers.
  • the VDU_A when it is determined that the VDU_A needs to be terminated, when the VDU of the type VDU_1 of the type VDU_A located on the physical server 1 is terminated, the VDU of the type VDU_3 of the type VDU_C located on the physical server 1 is also terminated.
  • steps S310-S311 may be replaced by the following processes:
  • Step S320 the network element management device receives the request for shrinking the VNF instance from the VNFM.
  • Step S321 the network element management device determines, according to the received contention VNF instance request and the resource configuration information stored by the network element management device, a VDU list and a VDU list that are required to be terminated by performing VNF instance shrinkage.
  • step S321 may include:
  • Step S3210 The network element management device determines, according to the VNF instance identifier and the capacity specification after the volume reduction VNF instance request, the VDU list to be terminated by performing the VNF instance shrinkage;
  • Step S3211 the network element management device is based on the number of bases of each VDU in the resource configuration information. According to the configuration information, the software to be deleted by each VDU in the VDU list to be terminated, the VM specification to be released, and the link related to the VM to be released are determined.
  • step S311 may further include step S3212, where the network element management device determines, according to the VM isolation and affinity in the resource configuration information, that the VDU in the VDU list to be terminated has The VDU of affinity and isolation.
  • the steps S3211-S3212 are the same as the steps S3111-S3112, respectively, and the specific implementations of the steps S3211-S3212 are not described herein.
  • Step S312 the network element management device performs VNF instance reduction according to the determined VDU list to be terminated and the VM resources to be released corresponding to each VDU in the VDU list.
  • step S312 includes:
  • Step S3121 The network element management device generates a command list for deleting the VDU according to the determined VDU list to be terminated. It is assumed that the list determined in step S311 includes one VDU of type A (simplified to: 1*VDU_A).
  • the embodiment of the present invention stores various types of configuration commands in the network element management device, for example, the deployment command ADD VDU, the deletion command RMV VDU, and the like. Therefore, in step S3120, the network element management device may generate a delete command for each VDU in the determined VDU list, and the delete command set generated by each VDU in the entire VDU list forms a VDU delete command list.
  • the determined VDU list is: 1*VDU_A
  • the embodiment of the present invention may generate the following VDU deletion command list:
  • VDU_1 (representing the number of the terminated VDU), VDU_A (representing the type of the terminated VDU);
  • step S3122 the network element management device uninstalls the software of each VDU in the process of executing the VDU deletion command in the command list of deleting the VDU, and requests the VIM to release the virtual machine VM resources related to each VDU through the VNFM. And deleting the basic data related to the VDU in the network element management device.
  • Step S3123 After performing all the VDU deletion commands in the command list of deleting the VDU, the network element management device notifies the VNFM that the VNF instantiation reduction is completed.
  • the network element management device obtains the performance parameter reported by the VNF, and determines, according to the reported performance parameter and the VNF instance shrinkage policy in the resource configuration information, that the execution of the VNF instance is to be terminated.
  • VDU list and a VM specification resource to be released corresponding to each VDU in the VDU list where the VDU list to be terminated includes one or more VDUs to be terminated, and the VNF instance shrinkage policy in the resource configuration information includes
  • the performance parameter of the last report is greater than the first threshold
  • the performance parameter of the current report is smaller than the difference between the first threshold and a buffer parameter, performing VNF instance shrinking, terminating one or more specified types of VDUs
  • the determined VDU list to be terminated and the VM resources to be released corresponding to the VDUs in the VDU list perform VNF instance reduction.
  • the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the lifecycle management of the VNF instance shrinkage does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to complete the corresponding actions step by step, avoiding the lifecycle management process of the VNF instance shrinkage. Mechanization and lack of flexibility.
  • the network element management device and the VNF usually belong to the same vendor. Therefore, the embodiment of the present invention can support the difference in the VNF instance shrinkage of each vendor, for example, the resource configuration stored in the network element management device of different vendors.
  • the format of the information may be different, and the specific processes of the network element management device of different vendors to perform VNF instance shrinkage according to the determined resources may also be different.
  • the ping-pong elastic expansion of the VNF is reduced by a buffer parameter.
  • FIG. 15 is a schematic flowchart diagram of a method for terminating a VNF instance according to an embodiment of the present invention.
  • the process of terminating the VNF instance in the embodiment of the present invention may be performed independently of the process of instantiating the VNF, or after the foregoing VNF instantiation process of the present invention. As shown in FIG. 15, it may include:
  • Step S410 The network element management device receives the terminating VNF instance request message sent by the VNF manager VMFM.
  • Step S411 the network element management device determines, according to the terminating VNF instance request message and the resource configuration information stored in the network element management device, a VDU list that needs to be terminated to terminate the VNF instance, and a corresponding VDU in the VDU list. VM resources to be released.
  • step S411 may specifically include:
  • Step S4110 the network element management device determines, according to the VNF instance identifier carried in the termination VNF instance request message and the resource configuration information stored in the network element management device, the VNF instance to be terminated and the VDU list to be terminated;
  • Step S4111 the network element management device determines, according to the basic data configuration information of each VDU in the resource configuration information, software that needs to be deleted for each VDU in the VDU list to be terminated, a VM specification to be released, and a release that needs to be released.
  • the link associated with the VM is the basic data configuration information of each VDU in the resource configuration information.
  • VM specifications corresponding to different types of VDUs may be different. Examples are as follows:
  • the VM specifications required for one VDU_A include: 1 VM_A, corresponding to 4 core vCPU + 8G memory + 2G bandwidth;
  • the VM specifications required for one VDU_B include: 1 VM_B, corresponding to 4 core vCPU + 16G memory + 2G bandwidth;
  • the VM specifications required for one VDU_C include: 1 VM_C, corresponding to 2 core vCPU + 8G memory + 2G bandwidth.
  • the embodiment of the present invention can determine the VM to be deleted according to the correspondence between the VDU and the VM specification. For example, when it is determined that the VDU list needs to be terminated as 1*VDU_A+1*VDU_B+1*VDU_C, it may be determined that one VM_A needs to be deleted, and the VM_A is deleted to release 4 cores vCPU+8G memory+2G bandwidth; VM_B, delete the VM_B specifically to release 4 core vCPU + 16G memory + 2G bandwidth; need to delete 1 VM_C, delete the VM_C specifically to release 2 core vCPU + 8G memory + 2G bandwidth.
  • the interworking relationship and bandwidth requirements of the network elements in the resource configuration information stored in the NE management device can be as follows:
  • VDU_A and VDU_B are interworking (that is, VM_A and VM_B are interworking), and the bandwidth needs to be 2G.
  • VDU_B is interworking with VDU_C (that is, VM_B and VM_C are interworking), and the bandwidth needs to be 2G.
  • step S4111 the embodiment of the present invention can determine that the link between VM_A and VM_B needs to be released according to the above-mentioned interworking relationship, and release the link between VM_B and VM_C.
  • step S411 the method further includes a step S4112, where the network element management device determines, according to the VM isolation and affinity in the resource configuration information, the list of VDUs to be terminated.
  • a VDU has a VDU with affinity and isolation relationships.
  • the VM isolation and affinity in the resource configuration information stored by the network element management device may be as follows:
  • VM_A and VM_C are deployed on the same physical server
  • the same type of VM is deployed on different physical servers.
  • the VDU_A when it is determined that the VDU_A needs to be terminated, when the VDU of the type VDU_1 of the type VDU_A located on the physical server 1 is terminated, the VDU of the type VDU_3 of the type VDU_C located on the physical server 1 is also terminated.
  • Step S412 the network element management device performs the VNF instance termination according to the determined VDU list to be terminated and the VM resources to be released corresponding to the VDUs in the VDU list.
  • step S412 includes:
  • Step S4121 The network element management device generates a command list for deleting the VDU according to the determined VDU list to be terminated. It is assumed that the list determined in step S411 is 1*VDU_A+1*VDU_B+1*VDU_C.
  • the network element management device of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a deletion command RMV VDU, and the like. Therefore, in step S4120, the network element management device may generate a delete command for each VDU in the determined VDU list, and the delete command set generated by each VDU in the entire VDU list forms a VDU delete command list.
  • the embodiment of the present invention may generate the following VDU deletion command list:
  • VDU_1 (representing the number of the terminated VDU), VDU_A (representing the type of the terminated VDU);
  • VDU_2 (representing the number of the terminated VDU), VDU_B (representing the type of the terminated VDU);
  • VDU_3 (representing the number of the terminated VDU), VDU_C (representing the type of the terminated VDU).
  • step S4122 the network element management device uninstalls the software of each VDU in the process of executing the VDU deletion command in the command list of deleting the VDU, and requests the VIM to be released through the VNFM.
  • Step S4123 After performing all VDU deletion commands in the command list of deleting the VDU, the network element management device notifies the VNFM that the VNF instantiation termination is completed.
  • the network element management device receives the termination VNF instance request message sent by the VNF manager VMFM, and determines the execution termination according to the terminating VNF instance request message and the resource configuration information stored in the network element management device.
  • the VNF instance needs to terminate the VDU list and the VM resources to be released corresponding to the VDUs in the VDU list, and the VDU list to be terminated includes one or more VDUs to be terminated, and the terminal to be terminated according to the determined
  • the VNF list and the VM resources to be released corresponding to the VDUs in the VDU list are executed, and the VNF instance is terminated.
  • the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the network element management device performs the lifecycle management of the VNF instance termination, it does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to complete the corresponding actions step by step, thereby avoiding the over-mechanization of the lifecycle management process of the VNF instance termination. Lack of flexibility.
  • the network element management device and the VNF are generally in the same vendor. Therefore, the embodiment of the present invention can support the difference in the termination of the VNF instance by each vendor, for example, the resource configuration information stored in the network element management device of different vendors.
  • the format of the VNF instance termination may be different, and the network management device of different vendors may perform the VNF instance termination according to the determined resources.
  • the embodiment of the present invention has the following advantages over the prior art VNFM (which can be the VNFM in the architecture shown in FIG. 1), which relies on the VNFD to perform the corresponding VNF lifecycle management:
  • VNFD provided by each Vendor is no longer required to cover all lifecycle management processes, nor does VNFD need to be standardized.
  • Vendor management devices of different Vendors can reflect the differences of Vendors.
  • the network element management device receives a lifecycle management operation, it is no longer necessary to strictly follow the corresponding parameters or processes of the VNFD and the instructions of the script to complete the corresponding step by step. The actions make the VNF lifecycle management process very flexible.
  • the final VNFD can be simplified as follows:
  • Simplified mode 1 Only the vendor (corresponding to the parameter of the parameter column in Table 1 and its corresponding parameter value), the VNF type (corresponding to the VNF id of the parameter column in Table 1 and its corresponding parameter value), and the version (correspondence table) are reserved in the VNFD.
  • Basic information such as the VNF version of the parameter column in 1 and its corresponding parameter value), the capacity specification (corresponding to the capacity of the parameter column in Table 1 and its corresponding parameter value).
  • Other information in the prior art VNFD does not need to be retained.
  • Simplified mode 2 The entire VNFD is removed, but the NSD template needs to be modified to ensure that the Vendor, VNF type, version information, and capacity specifications that can be obtained through the associated VNFD can be reflected in the NSD.
  • the NFVO->VNFM interface includes the following information as long as the method of the embodiment of the present invention is executed:
  • Vendor+VNF type used by the VNFM to forward the request to the correct NE management device
  • the VNF type, the VNF version number, and the capacity specification are forwarded by the VNFM to the network element management device, so that the network element management device uses this decision to instantiate which version of the VNF, and deploys according to the size.
  • the embodiment of the present invention needs to retain the parameters and scripts of the LMU in the VNFD or the NSD, so that the embodiment of the present invention performs VNF lifecycle management through the LMU.
  • the LMU could be deployed first.
  • the deployment process of the LMU is the same as that in the VNFD deployment process in the prior art, and is not described here.
  • the embodiment of the present invention further provides a network element management device that can be used to implement the foregoing methods.
  • the structural components of the network element management device in the embodiment of the present invention are described below with reference to the specific embodiments and the accompanying drawings.
  • FIG. 18 is a schematic structural diagram of an embodiment of a network element management device according to an embodiment of the present invention. As shown in FIG. 18, it may include: a receiving module 10, a determining module 11, and an instantiation module 12, where:
  • the receiving module 10 is configured to receive an instantiated VNF request message sent by the VNF manager VNFM.
  • the embodiment of the present invention may be in the network element.
  • An instance identifier is created in the management device for the VNF instance that is requested by the instantiated VNF request message, and the corresponding relationship between the version number, type, and capacity specification of the VNF instance and the instance identifier is established.
  • the network element management device and the VNF in the embodiment of the present invention are provided by the same provider, and may include, but are not limited to, an element management system (EMS) or a VNF local management unit (Local Management Unit). , LMU).
  • the receiving module 10 of the network element management device can receive an instantiated VNF request message from the VNFM through a Ve-Vnfm interface with the VNFM.
  • the determining module 11 is configured to determine, according to the instantiated VNF request message received by the receiving module and the resource configuration information stored in the network element management device, a resource required to perform the instantiation of the VNF.
  • the determining module 11 may specifically determine, according to the instantiated VNF request message and the resource configuration information stored in the network element management device, a virtual deployment unit VDU list required to instantiate the VNF, and the VDU list.
  • the virtual machine VM resource may include: a VM VM specification required by each VDU, a requirement of each VM to a link, and the like.
  • the resources required by the determining module 11 to perform the instantiation of the VNF may further include: software corresponding to each VDU and affinity and isolation of each VM.
  • the determining module 11 may include: a specification determining module 110, a VDU list determining module 111, a software determining module 112, a VM specification determining module 113, and a link determining module 114, where:
  • the specification determining module 110 is configured to determine a type, a version number, and a capacity specification of the VNF to be deployed according to the request of the instantiated VNF received by the receiving module 10.
  • the specification determining module 110 may describe the template VNFD index information and the capacity specification according to the information model of the VNF carried in the instantiated VNF request, and query the corresponding VNFD for the type, version number, and capacity specification of the VNF to be deployed. .
  • the version number and the capacity specification are shown in the method step S1110, and details are not described herein.
  • the specification determining module 110 may determine the type, version number, and capacity specification of the VNF to be deployed according to the type, version number, and capacity specification of the VNF carried by the instantiated VNF request. How to please The type, version number, and capacity specification of the VNF to be carried are determined. For example, refer to the method flow S1110, which is not described here.
  • the type, version number, and capacity specification of the VNF carried by the instantiated VNF request are registered in the VNFD or registered in an information model description template NSD of the network service.
  • the VDU list determining module 111 is configured to determine a VDU list required for the VNF to be deployed according to a ratio of a capacity specification in the resource configuration information to a virtual deployment unit VDU.
  • the specific means for the VDU list determining module 111 to determine the VDU list required for the VNF to be deployed may be the same as the step S1111, and details are not described herein.
  • the software determining module 112 is configured to determine, according to the correspondence between the VDU and the software in the resource configuration information, software corresponding to each VDU in the VDU list.
  • the VM specification determining module 113 is configured to determine, according to the correspondence between the VDU and the virtual machine (VM) in the resource configuration information, the VM specifications required by each VDU.
  • the VM specifications required for different types of VDUs may be different.
  • An example of the VM specification required for the VDU may be an example given in step S1113, and details are not described herein.
  • the link determining module 114 is configured to determine, according to the network interworking relationship in the resource configuration information and the bandwidth requirement of each link, a requirement of each VM for the link.
  • the means for the link determining module 114 to determine the respective requirements of the links of the VMs may be the same as the method step S1114, and details are not described herein.
  • the determining module 11 may further include an isolation affinity determining module 115, configured to determine the affinity and isolation of the VM according to the VDU isolation and affinity in the resource configuration information. .
  • the means for the isolation affinity determination module 115 to determine the respective requirements of the respective VMs may be the same as the method step S1115, and details are not described herein.
  • the instantiation module 12 is configured to instantiate the VNF according to the resource determined by the determining module.
  • the instantiation module 12 may specifically include: a generating module 120, a deployment module 121, and a notification module 122, where:
  • the generating module 120 is configured to generate a command list for deploying the VDU according to the determined VDU list.
  • the network element management device of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a deletion command RMV VDU, and the like. Therefore, the generating module 120 may generate a deployment command for each VDU in the determined VDU list, and the deployment command set generated by each VDU in the entire VDU list forms a VDU deployment command list.
  • An example of the VDU deployment command list that can be generated in this embodiment may be an example in the step S1120, and details are not described herein.
  • the deployment module 121 is configured to execute the VDU deployment command in the command list of the deployment VDU generated by the generation module 120, and request the VIM to request the command list in the VDU deployment command in the command list.
  • the virtual machine VM resources required by the VDU are loaded with software and basic data for each VDU in the VDU list according to the allocation result of the VIM to the virtual machine VM resources required by each VDU in the command list.
  • the deployment module 121 may allocate a corresponding resource to the VIM request according to the VM resource and the link resource determined by the determining module 11, and load software and perform basic data configuration for each VDU in the VDU list.
  • the performing the basic data configuration means that the software loaded for each VDU, the allocation result of the virtual machine VM resources allocated by each VDU, and the type of each VDU are recorded in the network element management device, and Each VDU is associated with an identifier of a VNF instance including the VDUs as a subsequent available resource configuration information, so that the network element management device manages the VNFs in the VNF instance and the instance based on the basic data configuration.
  • the deployment module 121 may execute the deployment command of each VDU in the command list one by one, and in the process of executing the deployment command of each VDU, request the VIM to allocate the required VDU.
  • a virtual machine VM resource and according to the resource allocated by the VIM for each VDU, loading software for the one VDU and allocating a required virtual machine VM resource allocation result to each VDU according to the VIM,
  • Each VDU loads the software and performs basic data configuration.
  • the deployment module 121 may also request the VM resources and link resources required by all the VDUs in the VDU list to the VIM at one time, and load the software and perform basic data for each VDU at one time after the request is completed. Configuration.
  • the notification module 122 is configured to notify the VNFM that the instantiated VNF is completed after the deployment module 121 executes the deployment command of all the VDUs in the VDU list.
  • the network element management device of the embodiment of the present invention may further include an operation authorization processing module, configured to virtualize an orchestrator to the network function via a VNFM (VNFM) ( Network Functions Virtualization Orchestrator (NFVO) initiating an operation authorization request to instantiate a VNF, and receiving, by the VNFM, a response to an operation authorization of the instantiated VNF from the NFVO, the response of the operational authorization of the instantiated VNF is NFVO determines that there is a resource required to execute the instantiated VNF and requests the VIM to reserve a corresponding resource and then send it;
  • VNFM VNFM
  • NFVO Network Functions Virtualization Orchestrator
  • the deployment module 121 requests the VIM to allocate VM resources and link resources required for each VDU in the VDU list, specifically for allocating the request to each VDU in the VDU list via the VNFM.
  • the request for the VM resource is forwarded to the VIM.
  • the deployment module 121 requests the VIM to allocate VM resources required for each VDU in the VDU list, specifically for sequentially allocating requests to each VDU in the VDU list via VNFM and NFVO.
  • a request for the required VM resource is forwarded to the VIM.
  • the NFVO directly initiates a resource allocation request to the VIM, and the previous embodiment further includes an operation authorization request and a response to determine that there are related resources in the network before initiating the resource allocation request. It can be seen that the network element management device in the embodiment of the present invention transfers the lifecycle management work of the VNFM to the network element management device.
  • the VNFD does not need to set a complete parameter that can cover each VNF lifecycle management process or script.
  • the network element management device does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to perform the lifecycle management of the VNF, and the corresponding actions are completed step by step, thereby avoiding the over-mechanization of the lifecycle management process of the instantiated VNF. Lack of flexibility.
  • the network element management device and the VNF usually belong to the same provider. Therefore, the embodiment of the present invention can support the difference in the performance of the VNF generated by each vendor, for example, the resource configuration information stored in the network element management device of different vendors.
  • the format may be different, and the specific processes of the network element management device of different vendors to instantiate the VNF according to the determined resources may also be different.
  • the network element management device of this embodiment may further include:
  • the determining module 11 is further configured to determine, according to the reported performance parameter and the VNF instance shrinking policy in the resource configuration information, a VDU list to be terminated by performing VNF instance shrinkage and a corresponding VDU in the VDU list.
  • the VM resource to be released, the VDU list to be terminated includes one or more VDUs to be terminated, and the VNF instance shrinkage policy in the resource configuration information includes: when the performance parameter of the last report is greater than the first threshold, When the performance parameter reported in this time is smaller than the difference between the first threshold and a buffer parameter, the VNF instance is compacted to terminate one or more specified types of VDUs.
  • the network element management device of this embodiment may further include: a volume reduction module (not shown), configured to determine, according to the determining module 11, the VDU list to be terminated and the VM to be released corresponding to each VDU in the VDU list. Resources, perform VNF instance shrinkage.
  • a volume reduction module (not shown), configured to determine, according to the determining module 11, the VDU list to be terminated and the VM to be released corresponding to each VDU in the VDU list. Resources, perform VNF instance shrinkage.
  • FIG. 21 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention. As shown in FIG. 21, it may include: an obtaining module 20, a determining module 21, and a expansion module 22, wherein:
  • the obtaining module 20 is configured to obtain performance parameters reported by the VNF.
  • the performance parameter of the VNF reported by the obtaining module 20 of the embodiment of the present invention may include a CPU usage rate, or any other performance parameter that can be reported by the VNF.
  • the determining module 21 is configured to determine, according to the performance parameter reported by the VNF and the resource configuration information stored in the network element management device, a VDU list required for performing VNF instance expansion and a VDU required for each VDU in the VDU list. Virtual machine VM resources.
  • the embodiment of the present invention may maintain a VNF instance expansion and contraction policy information in the resource configuration information stored in the network element management device, or maintain a VNF instance expansion strategy information in the resource configuration information.
  • a VNF instance shrinkage policy information where the VNF instance expansion policy may include: when the performance parameter of the last report is lower than the second threshold, and the performance parameter of the current report is greater than or equal to the second threshold, executing the VNF instance
  • the capacity expansion is performed by deploying one or more specified types of VDUs based on the existing VNF instance capacity specifications.
  • the VNF instance shrinking policy may include: when the last reported performance parameter is greater than the first threshold, the present report is performed. When the performance parameter is less than the difference between the first threshold and a buffer parameter, performing VNF instance shrinking, terminating one or more specified types of VDUs in the VNF instance.
  • VNF expansion and contraction policy and the performance parameter of the VNF which are stored in the embodiment of the present invention, may be referred to the related description in step S211, and details are not described herein.
  • the determining module 21 may determine the VDU list to be deployed for performing the VNF instance expansion, for example, when the foregoing policy is met. At 1 o'clock, it is determined on the basis of the original instance of the VNF that a VDU of type A needs to be deployed.
  • the resource configuration information stored in the network element management device may further include a ratio of the capacity specification to the virtual deployment unit VDU, a correspondence between the VDU and the software, and a corresponding relationship between the VDU and the virtual machine. Or information about network interworking relationships and bandwidth requirements of each link. Therefore, in the embodiment of the present invention, after determining the VDU list to be deployed, the virtual machine VM resources required for each VDU in the VDU list may be further determined according to the resource configuration information stored in the network element management device.
  • the virtual machine VM resource may include: a VM VM specification required by each VDU, a requirement of each VM to a link, and the like.
  • the resources determined by the embodiments of the present invention may further include: software corresponding to each VDU and affinity and isolation of each VM.
  • the determining module 21 of the embodiment of the present invention may specifically include: a VDU list determining module 211, a software determining module 212, a VM specification determining module 213, and a link determining module 214, where:
  • the VDU list determining module 211 is configured to determine, according to the performance parameter that is reported by the VNF and the VNF instance expansion policy information in the resource configuration information stored in the network element management device, a VDU list that needs to be deployed to perform VNF instance expansion. Specifically, how to determine the VDU list refers to the previous description, and details are not described herein.
  • the software determining module 212 is configured to determine, according to the correspondence between the VDU and the software in the resource configuration information stored in the network element management device, the software corresponding to each VDU in the VDU list that is required to be deployed in the VNF instance expansion.
  • the VM specification determining module 213 is configured to determine, according to the mapping between the VDU and the virtual machine (VM) in the resource configuration information stored in the network element management device, the VDU list to be deployed in the VNF instance expansion.
  • the VM specifications required by each VDU In the specific implementation, different The VM specifications required for a type of VDU may vary. An example of the correspondence between the VDU and the VM specification can be seen in the description of the step S2112, and details are not described herein.
  • the link determining module 214 is configured to: according to the network interworking relationship in the resource configuration information stored in the network element management device and the bandwidth requirement of each link, the VMs respectively require a link.
  • the example of the interworking relationship and the bandwidth requirement of the network element in the resource configuration information stored in the network element management device may be as described in step S2113, and details are not described herein.
  • the determining module 21 may further include an isolation affinity determining module 215, configured to determine, according to the VDU isolation and affinity in the resource configuration information stored in the network element management device.
  • VM affinity and isolation For example, the example of the VDU isolation and the affinity relationship in the resource configuration information stored by the network element management device may be as described in step S2114, and details are not described herein.
  • the isolation affinity determination module 215 can determine the affinity and isolation of each VM according to the VDU isolation and affinity stored by the network element management device.
  • VM_A and VM_C are deployed on the same physical server, and multiple VM_A deployments are performed. On different physical servers, for example, the types of VM_1 and VM_2 are VM_A, but when deployed, VM_1 and VM_2 need to be deployed on different physical servers.
  • the expansion module 22 may specifically include:
  • the generating module 221 is configured to generate a command list for deploying the VDU according to the VDU list determined by the determining module 21. It is assumed that the VDU list determined in step S211 or S221 includes one VDU of type A (simplified as: 1*VDU_A).
  • the network element management device of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a deletion command RMV VDU, and the like. Therefore, the generating module 221 of the network element management device may generate a deployment command for each VDU in the determined VDU list, and the deployment command set generated by each VDU in the entire VDU list forms a VDU deployment command list.
  • the example of the VDU deployment command list generated in this embodiment may be the same as the example in step S2120, and details are not described herein.
  • the deployment module 222 is configured to execute a deployment command of the VDU in the command list, and in the process of executing the deployment command of the VDU in the command list, request the VIM in the command list from the VIM.
  • the virtual machine VM resources are required, and the software is loaded and the basic data is configured for each VDU in the VDU list according to the allocation result of the VIM to the virtual machine VM resources required by each VDU in the command list.
  • the performing the basic data configuration means that the software loaded for each VDU and the allocation result of the virtual machine VM resources allocated by each VDU and the type of each VDU are recorded in the network element management device for subsequent use. Resource configuration information, and associates each VDU with the identifier of the VNF instance including the VDUs as the subsequent available resource configuration information, so that the network element management device performs the VNF in the VNF instance and the instance based on the basic data configuration. management. It can be seen that the basic data configuration is automatically completed in the process of performing the VNF instance expansion in the embodiment of the present invention, which solves the technical problem that the prior art needs to manually perform basic data configuration in the network element management device.
  • the deployment module 222 may execute the deployment command of each VDU in the command list one by one, and in the process of executing the deployment command of each VDU, request the VIM to be allocated for each VDU.
  • a virtual machine VM resource and according to the resource allocated by the VIM for each VDU, loading software for the one VDU and allocating a required virtual machine VM resource allocation result to each VDU according to the VIM,
  • Each VDU loads the software and performs basic data configuration.
  • the deployment module 222 may also request the VM resources and link resources required by all the VDUs in the VDU list to the VIM at one time, and load the software and perform basic data for each VDU once after the request is completed. Configuration.
  • the notification module 223 is configured to notify the VNFM that the VNF instance expansion is completed after the deployment module 222 executes the deployment command of all the VDUs in the VDU list.
  • the embodiment of the present invention may further include:
  • An operation authorization processing module (not shown), configured to initiate an operation authorization request of the VNF instance expansion to the network function virtualization orchestrator NFVO via the VNF manager VNFM, and receive the VNF instance expansion from the NFVO via the VNFM.
  • the response of the operation authorization of the VNF instance expansion is that the NFVO determines that there is a resource required to perform the expansion of the VNF instance and requests the VIM to reserve a corresponding resource.
  • the deployment module 222 requests the VIM to allocate each VDU in the VDU list.
  • the VM resource is specifically used to forward, by the VNFM, a request for allocating VM resources required for each VDU in the VDU list to the VIM.
  • the deployment module 222 when the deployment module 222 requests the VIM to allocate the VM resources required for each VDU in the VDU list, the deployment module 222 specifically allocates the VDUs in the VDU list through the VNFM and the NFVO. A request for the required VM resource is forwarded to the VIM.
  • the NFVO directly initiates a resource allocation request to the VIM, and the previous embodiment further includes an operation authorization request and a response to determine that there are related resources in the network before initiating the resource allocation request.
  • the network element management device of the embodiment of the present invention may include: a receiving module 23, a determining module 24, and a capacity expansion module 22, wherein the expansion module 22 is FIG. 21 or The expansion module in Figure 23 is not described here, among them,
  • the receiving module 23 is configured to receive a VNF instance expansion request from the VMFM.
  • the determining module 24 is configured to determine, according to the VNF instance expansion request received by the receiving module 23 and the resource configuration information stored by the network element management device, a VDU list that needs to be deployed to perform VNF instance expansion, and each VDU in the VDU list.
  • the required virtual machine VM resources are configured to determine, according to the VNF instance expansion request received by the receiving module 23 and the resource configuration information stored by the network element management device, a VDU list that needs to be deployed to perform VNF instance expansion, and each VDU in the VDU list. The required virtual machine VM resources.
  • the determining module 24 may include: a VDU list determining module 241, a software determining module 242, a VM specification determining module 243, and a link determining module 244, where:
  • the VDU list determining module 241 is configured to determine a VDU list to be deployed according to the VNF instance identifier carried in the VNF instance expansion request and the capacity specification after the expansion;
  • the software determining module 242 is configured to determine, according to the corresponding relationship between the VDU and the software in the resource configuration information stored by the network element management device, software corresponding to each VDU in the VDU list that is required to be deployed in the VNF instance expansion;
  • the VM specification determining module 243 is configured to determine, according to the correspondence between the VDU and the virtual machine (VM) in the resource configuration information stored by the network element management device, each VDU in the VDU list that needs to be deployed to perform VNF instance expansion.
  • the link determining module 244 is configured to: according to the network in the resource configuration information stored by the network element management device The interworking relationship and the bandwidth requirements of each link, the respective requirements of the VM for the link.
  • the determining module 24 may further include an isolation affinity determining module 245, configured to determine, according to the VDU isolation and affinity in the resource configuration information stored by the network element management device. VM affinity and isolation.
  • the software determining module 242, the VM specification determining module 243, the link determining module 244, and the isolation affinity determining module 245 are respectively determined with the software determining module 212, the VM specification determining module 213, the link determining module 214, and the isolation affinity. Modules 215 are identical, and therefore, their specific implementations are not described herein.
  • the embodiment of the present invention transfers the lifecycle management work of the VNFM to the network element management device. Therefore, the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the lifecycle management of the VNF instance expansion does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to complete the corresponding actions step by step. This avoids the over-mechanization of the lifecycle management process of the VNF instance expansion. Lack of flexibility.
  • the network element management device and the VNF are generally the same vendor. Therefore, the embodiment of the present invention can support the difference in the VNF instance expansion of each vendor, for example, the resource configuration information stored in the network element management device of different vendors.
  • the format of the VNF instance expansion may be different according to the determined resources.
  • FIG. 26 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention. As shown in FIG. 26, it may include: an obtaining module 30, a determining module 31, and a shrinking module 32, wherein:
  • the obtaining module 30 is configured to obtain performance parameters reported by the VNF.
  • the performance parameter reported by the VNF obtained by the embodiment of the present invention may include a CPU usage rate, or any other performance parameter that can be reported by the VNF.
  • a determining module 31 configured to determine, according to the reported performance parameter acquired by the obtaining module 30 and the VNF instance shrinking policy in the resource configuration information, a VDU list to be terminated by performing VNF instance shrinkage and each of the VDU lists
  • the VDU corresponding to the VM resource to be released, the VDU list to be terminated includes one or more VDUs to be terminated
  • the VNF instance shrinkage policy in the resource configuration information includes: when the performance parameter of the last report is greater than the first A threshold value, the performance parameter reported this time is smaller than the first When the threshold is different from a buffered parameter, the VNF instance is compacted to terminate one or more VDUs of the specified type.
  • the embodiment of the present invention may maintain a VNF instance expansion and contraction policy information in the resource configuration information stored in the network element management device, or maintain a VNF instance expansion strategy information in the resource configuration information.
  • a VNF instance shrinkage policy information where the VNF instance expansion policy may include: when the performance parameter of the last report is lower than the second threshold, and the performance parameter of the current report is greater than or equal to the second threshold, executing the VNF instance
  • the capacity expansion is performed by deploying one or more specified types of VDUs based on the existing VNF instance capacity specifications.
  • the VNF instance shrinking policy may include: when the last reported performance parameter is greater than the first threshold, the present report is performed. When the performance parameter is less than the difference between the first threshold and a buffer parameter, performing VNF instance shrinking, terminating one or more specified types of VDUs in the VNF instance.
  • the description of the VNF expansion and contraction policy and the performance parameter of the VNF stored in this embodiment may be referred to the related description of step S311, and details are not described herein.
  • the determining module 31 may determine a VDU list that needs to be terminated to perform the VNF instance shrinking, for example, when When the above strategy 3 is satisfied, it is determined that a VDU of type A needs to be terminated on the basis of the original instance of the VNF.
  • the resource configuration information stored in the network element management device may further include a ratio of the capacity specification to the virtual deployment unit VDU, a correspondence between the VDU and the software, and a corresponding relationship between the VDU and the virtual machine. Information such as network interworking relationships and bandwidth requirements of each link. Therefore, in the embodiment of the present invention, after the determining module 31 determines the VDU list to be terminated, the virtual machine VM resources to be released by the VDUs in the VDU list may be further determined according to the resource configuration information stored in the network element management device. And the software that each VDU needs to delete.
  • the virtual machine VM resource may include: a VM VM specification required by each VDU, a requirement of each VM to a link, and the like. Further, as a supplementary embodiment, the determining module 31 of the embodiment of the present invention may further determine a VDU that has a relationship of affinity and isolation with the determined VDU to be terminated.
  • the determining module 31 of the embodiment of the present invention may specifically include: a VDU list determining module 311 and a release resource determining module 312, where:
  • the VDU list determining module 311 is configured to determine, according to the performance parameter of the VNF report obtained by the obtaining module 30 and the VNF instance shrinkage policy information in the resource configuration information stored in the network element management device, that the VNF instance is reduced. A list of VDUs to terminate. Specifically, how to determine the previous description is not described here.
  • the release resource determining module 312 is configured to determine, according to the basic data configuration information of each VDU in the resource configuration information, software that needs to be deleted in each VDU in the VDU list to be terminated, a VM specification to be released, and a release to be released. VM related link.
  • the basic data configuration is performed by recording the software loaded for each VDU and the allocation result of the virtual machine VM resources allocated by each VDU and the type of each VDU into the network element management device as a subsequently available resource.
  • the configuration information is such that the network element management device performs subsequent management of the VDU based on the basic data configurations. Therefore, the basic data configuration of each VDU includes information such as a corresponding VM specification of the VDU, software corresponding to the VDU, and a link of the VM corresponding to the VDU.
  • VM specifications corresponding to different types of VDUs may be different.
  • An example of the correspondence between the VDU and the VM specification may be an example of the step S3111, and details are not described herein.
  • the example of the interworking relationship and the bandwidth requirement of the network element in the resource configuration information stored in the network element management device in this embodiment may be as shown in step S3111, and details are not described herein.
  • the release resource determining module 312 may further determine, according to the foregoing interworking relationship, that a link between VM_A and VM_B needs to be released, and release a link between VM_B and VM_C.
  • the resource release determining module 321 may also delete the basis of the VDU configured in the network element management unit.
  • the data for example, the association relationship between the VDU and the VNF instance is deleted, the stored correspondence information between the VDU and the software is deleted, the correspondence between the VDU and the VM specification is deleted, and the VM of the VDU is deleted. Link information, etc.
  • the resource release determining module 312 is further configured to determine, according to VM isolation and affinity in the resource configuration information, affinity and isolation with the VDU in the VDU list that needs to be terminated. VDU of degree relationship.
  • the network element management device of the embodiment stores An example of VM isolation and affinity in the resource configuration information may be shown in step S31112, and details are not described herein.
  • the VDU_A when it is determined that the VDU_A needs to be terminated, when the VDU of the type VDU_1 of the type VDU_A located on the physical server 1 is terminated, the VDU of the type VDU_3 of the type VDU_C located on the physical server 1 is also terminated.
  • the volume reduction module 32 is configured to perform VNF instance reduction according to the VDU list to be terminated determined by the determining module 31 and the VM resources to be released corresponding to the VDUs in the VDU list.
  • the volume reduction module 32 of the embodiment of the present invention specifically includes:
  • the generating module 321 is configured to generate a command list for deleting the VDU according to the VDU list to be terminated determined by the determining module 31.
  • the list determined by the determining module 31 includes one VDU of type A (simplified to: 1*VDU_A).
  • the network element management device of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a deletion command RMV VDU, and the like. Therefore, the generating module 321 can generate a delete command for each VDU in the determined VDU list, and the delete command set generated by each VDU in the entire VDU list forms a VDU delete command list.
  • the determined VDU list is: 1*VDU_A
  • the embodiment of the present invention may generate the following VDU deletion command list:
  • VDU_1 (representing the number of the terminated VDU), VDU_A (representing the type of the terminated VDU);
  • the deleting module 322 is configured to execute the VDU delete command in the command list of the deleted VDU, and uninstall the software of each VDU in the process of executing the VDU delete command in the command list of deleting the VDU, and use the VNFM to The VIM requests to release virtual machine VM resources related to each VDU, and deletes the basic data related to the VDU in the network element management device.
  • the notification module 323 is configured to notify the VNFM that the VNF instantiation reduction is completed after the deleting module 322 executes all the VDU deletion commands in the command list for deleting the VDU.
  • the embodiment of the present invention transfers the lifecycle management work of the VNFM to the network element management device. Therefore, the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the NE management device does not need to perform lifecycle management of VNF instance shrinkage.
  • the corresponding actions are completed step by step, which avoids the over-mechanization and lack of flexibility of the lifecycle management process of VNF instance shrinkage.
  • the network element management device and the VNF usually belong to the same vendor. Therefore, the embodiment of the present invention can support the difference in the VNF instance shrinkage of each vendor, for example, the resource configuration stored in the network element management device of different vendors.
  • the format of the information may be different, and the specific processes of the network element management device of different vendors to perform VNF instance shrinkage according to the determined resources may also be different.
  • the ping-pong elastic expansion of the VNF is reduced by a buffer parameter.
  • the network element management device of the embodiment of the present invention may include: a receiving module 33, a determining module 34, and a shrinking module 32, wherein the volume reducing module 32 and FIG. 28 The same as the reduction module in Figure 26, and will not be described here, where:
  • the receiving module 33 is configured to receive a VNF instance shrink request from the VNFM.
  • the determining module 34 is configured to determine, according to the shrinking VNF instance request received by the receiving module 33 and the resource configuration information stored by the network element management device, a VDU list and a VDU list that are required to be terminated by performing VNF instance shrinking.
  • the determining module 34 may include: a VDU list determining module 341 and a release resource determining module 342, where:
  • the VDU list determining module 341 is configured to determine, according to the VNF instance identifier and the capacity specification after the volume reduction VNF instance request received by the receiving module 33, the VDU list to be terminated by performing the VNF instance shrinkage;
  • the release resource determining module 342 is configured to determine, according to the basic data configuration information of each VDU in the resource configuration information, software that needs to be deleted for each VDU in the VDU list to be terminated, a VM specification to be released, and a release to be released. VM related link.
  • the release resource determining module 342 is further configured to determine, according to VM isolation and affinity in the resource configuration information, affinity and isolation with the VDU in the VDU list that needs to be terminated. VDU of degree relationship.
  • the release resource determination module 342 of the present embodiment is the same as the release resource determination module 322 of the previous embodiment, and details are not described herein.
  • FIG. 31 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention. As shown in FIG. 31, it may include: a receiving module 40, a determining module 41, and a terminating module 42, wherein:
  • the receiving module 40 is configured to receive a termination VNF instance request message sent by the VNF manager VMFM.
  • the determining module 41 is configured to determine, according to the terminating VNF instance request message received by the receiving module 40 and the resource configuration information stored in the network element management device, a VDU list that needs to be terminated to terminate the VNF instance, and each VDU in the VDU list. Corresponding VM resources to be released.
  • the determining module 41 may specifically include: a VDU list determining module 411 and a release resource determining module 412, where:
  • the VDU list determining module 411 is configured to determine, according to the VNF instance identifier carried in the terminating VNF instance request message and the resource configuration information stored in the network element management device, a VNF instance to be terminated and a VDU list to be terminated;
  • the release resource determining module 412 is configured to determine, according to the basic data configuration information of each VDU in the resource configuration information, software that needs to be deleted for each VDU in the VDU list to be terminated, a VM specification to be released, and a release to be released. VM related link.
  • the basic data configuration is performed by recording the software loaded for each VDU and the allocation result of the virtual machine VM resources allocated by each VDU and the type of each VDU into the network element management device as a subsequently available resource.
  • the configuration information is such that the network element management device performs subsequent management of the VDU based on the basic data configurations. Therefore, the basic data configuration of each VDU includes information such as a corresponding VM specification of the VDU, software corresponding to the VDU, and a link of the VM corresponding to the VDU.
  • VM specifications corresponding to different types of VDUs may be different.
  • an example of the correspondence between the VDU and the VM specification in this embodiment may be an example in the step S4111, and details are not described herein.
  • the release resource determining module 412 of the embodiment of the present invention can determine the VM to be deleted according to the correspondence between the VDU and the VM specification. For example, when it is determined that the VDU list needs to be terminated as 1*VDU_A+1*VDU_B+1*VDU_C, it may be determined that one VM_A needs to be deleted, and the VM_A is deleted to release 4 cores vCPU+8G memory+2G bandwidth; VM_B, delete the VM_B specifically to release 4 core vCPU + 16G memory + 2G bandwidth; need to delete 1 VM_C, delete the VM_C is specifically required to release 2 cores vCPU + 8G memory + 2G bandwidth.
  • the example of the interworking relationship and the bandwidth requirement of the network element in the resource configuration information stored in the network element management device in this embodiment may be as shown in step S4111, and details are not described herein.
  • the release resource determining module 412 of the embodiment of the present invention can determine that the link between VM_A and VM_B needs to be released according to the above-mentioned interworking relationship, and release the link between VM_B and VM_C.
  • the release resource determining module 412 may further determine, according to the VM isolation and affinity in the resource configuration information, the affinity and isolation relationship with the VDU in the VDU list that needs to be terminated. VDU.
  • the example of the VM isolation and the affinity in the resource configuration information stored in the network element management device of the embodiment may be as shown in step S41112, and details are not described herein.
  • the termination module 42 is configured to perform the VNF instance termination according to the VDU list to be terminated determined by the determining module 41 and the VM resources to be released corresponding to the VDUs in the VDU list.
  • the termination module 42 specifically includes:
  • the generating module 421 is configured to generate a command list for deleting the VDU according to the determined VDU list to be terminated. As an embodiment, it is assumed that the list determined by the determining module 41 is 1*VDU_A+1*VDU_B+1*VDU_C.
  • the network element management device of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a deletion command RMV VDU, and the like. Therefore, the generating module 421 can generate a delete command for each VDU in the determined VDU list, and the delete command set generated by each VDU in the entire VDU list forms a VDU delete command list.
  • the deleting module 422 is configured to execute the VDU delete command in the command list for deleting the VDU, and uninstall the software of each VDU in the process of executing the VDU delete command in the command list of deleting the VDU, and use the VNFM to The VIM requests to release virtual machine VM resources related to each VDU, and deletes the basic data related to the VDU in the network element management device.
  • a notification module 423 configured to execute the delete command list of the deleted VDU in the deleting module 422 After all the VDUs in the delete command, the VNFM is notified that the VNF instantiation is terminated.
  • the embodiment of the present invention transfers the lifecycle management work of the VNFM to the network element management device. Therefore, the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the network element management device performs the lifecycle management of the VNF instance termination, it does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to complete the corresponding actions step by step, thereby avoiding the over-mechanization of the lifecycle management process of the VNF instance termination. Lack of flexibility.
  • the network element management device and the VNF are generally in the same vendor.
  • the embodiment of the present invention can support the difference in the termination of the VNF instance by each vendor, for example, the resource configuration information stored in the network element management device of different vendors.
  • the format of the VNF instance termination may be different, and the network management device of different vendors may perform the VNF instance termination according to the determined resources.
  • the above is a description of the functional modules of the network element management device.
  • the following is an example of the hardware components of the network element management device and the division of labor of each part of the hardware.
  • FIG. 34 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention. As shown in FIG. 34, it may include: an input device 50, an output device 51, a communication link 52, a transceiver device 53, a memory 54, and a processor 55, wherein:
  • the input device 50 is configured to receive input data externally input to the network element management device.
  • the input device 50 may include a keyboard, a mouse, a photoelectric input device, a sound input device, a touch input device, a scanner, and the like.
  • the input data can include various user instructions.
  • the output device 51 is configured to output the output data of the network element management device.
  • the output device 51 may include a display, a speaker, a printer, and the like.
  • the output data can include response data to user instructions.
  • the communication link 52 is configured to communicate with other devices.
  • the communication link 52 described in the embodiment of the present invention may be an example of a propagation medium.
  • the propagation medium can generally embody computer readable instructions, data structures, program modules, or other data in the form of other modulated data signals, such as a carrier wave or other transport mechanism.
  • the communication medium can include wired media, such as a priority network or In a straight line connection, the propagation medium may also include a wireless medium such as sound waves, radio frequency, infrared rays, and the like.
  • the communication link 52 of the embodiment of the present invention may be used to transmit a VNF instantiation request message and transmit a message of the instantiation VNF completion sent to the VNFM.
  • the transceiver device 53 is configured to send data to or receive data from other devices through the communication link 52.
  • the transceiver device 53 of the embodiment of the present invention may be configured to receive a VNF instantiation request message and send a message to the VNFM to instantiate the VNF completion.
  • the transceiver device 53 can be a Ve-Vnfm interface with the VNFM.
  • the memory 54 is configured to store program data with various functions.
  • the data stored in the memory 54 in the embodiment of the present invention includes resource configuration information, and program data called and executed by the processor.
  • the memory 54 of an embodiment of the present invention may be a system memory such as volatile (such as RAM), non-volatile (such as ROM, flash memory, etc.), or a combination of both.
  • the memory 54 of the embodiment of the present invention may also be an external memory outside the system, such as a magnetic disk, an optical disk, a magnetic tape, or the like.
  • the processor 55 is configured to invoke program data and resource configuration information stored in the memory 54 and perform the following operations:
  • A1 The instantiated VNF request message sent by the VNF manager VNFM is received from the transceiver 53.
  • A2 Determine the resources required to perform the instantiation of the VNF according to the instantiated VNF request message and the resource configuration information stored in the memory 54.
  • A3 Instantiate the VNF according to the determined resource.
  • the network element management device and the VNF in the embodiment of the present invention are provided by the same provider, and may include, but are not limited to, an element management system (EMS) or a VNF local management unit (Local Management Unit). , LMU).
  • EMS element management system
  • LMU VNF local management unit
  • the processor 55 may create an instance identifier for the VNF instance that is requested to be established by the instantiated VNF request message in the network element management device, and establish a version of the VNF instance. The correspondence between the number, type, and capacity specifications and the instance ID.
  • the processor 55 may determine, according to the instantiated VNF request message and the resource configuration information stored in the network element management device, a virtual deployment unit VDU list required to instantiate the VNF, and Virtual machine VM resources required for each VDU in the VDU list.
  • the virtual machine VM resource may include: a VM VM specification required by each VDU, a requirement of each VM to a link, and the like.
  • the resources required to perform the instantiation of the VNF determined by the embodiment of the present invention may further include: software corresponding to each VDU and affinity and isolation of each VM.
  • the processor 55 determines the resources required to perform the instantiation of the VNF according to the instantiated VNF request message and the resource configuration information stored in the memory 54, the following operations are specifically performed:
  • A21 Determine, according to the request for instantiating the VNF, a type, a version number, and a capacity specification of the VNF to be deployed.
  • the processor 55 may describe the template VNFD index information and the capacity specification according to the information model of the VNF carried in the instantiated VNF request, and query the corresponding VNFD for the type, version number, and capacity specification of the VNF to be deployed. For details on how to find the corresponding VNFD through the VNFD index and the type of the VNF to be deployed, the version number and the capacity specification are shown in the method step S1110, and details are not described herein.
  • the processor 55 may determine the type, version number, and capacity specification of the VNF to be deployed according to the type, version number, and capacity specification of the VNF carried by the instantiated VNF request. For an example of how to configure the VNF type, version number, and capacity specification of the VNF to be deployed, refer to the method flow S1110, which is not described here.
  • the type, version number, and capacity specification of the VNF carried by the instantiated VNF request are registered in the VNFD or registered in an information model description template NSD of the network service.
  • A22 Determine, according to a ratio of the capacity specification in the resource configuration information to the virtual deployment unit VDU, a VDU list required for the VNF to be deployed.
  • the specific means for the processor 55 to determine the VDU list required for the VNF to be deployed may be the same as the step S1111, and details are not described herein.
  • A23 Determine, according to the correspondence between the VDU and the software in the resource configuration information, software corresponding to each VDU in the VDU list.
  • A24 Determine, according to the correspondence between the VDU and the virtual machine (VM) in the resource configuration information, the VM specifications required by each VDU.
  • VM virtual machine
  • the required VM specifications can be different.
  • An example of the VM specification required for the VDU may be an example given in step S1113, and details are not described herein.
  • A25 Determine, according to the network interworking relationship in the resource configuration information and the bandwidth requirement of each link, the requirements of each VM for the link.
  • the interworking relationship and the bandwidth requirement of the network element in the resource configuration information stored in the memory 54 can be referred to the example shown in step S1114, and details are not described herein.
  • the link requirement determined by the processor 55 may be a virtual link between VM_A and VM_B, and a virtual link needs to be established between VM_B and VM_C.
  • the processor 55 further determines the affinity and isolation of the VM according to the VDU isolation and affinity in the resource configuration information.
  • the VDU isolation and affinity information in the resource configuration information stored in the memory 54 may be the same as the example in the step S1115, and details are not described herein.
  • the processor 55 can determine the affinity and isolation of each VM according to the VDU isolation and affinity stored by the network element management device.
  • VM_A and VM_C are deployed on the same physical server, and multiple VM_As are deployed on different physical servers.
  • VM_1 and VM_2 are both of type VM_A, but when deployed, VM_1 and VM_2 need to be deployed on different physical servers.
  • the processor 55 when the processor 55 performs the instantiation of the VNF according to the determined resource, the following steps may be specifically performed:
  • A31 Generate a command list for deploying the VDU according to the determined VDU list.
  • the memory 54 of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a delete command RMV VDU, and the like. Therefore, the processor 55 can generate a deployment command for each VDU in the determined VDU list, and the deployment command set generated by each VDU in the entire VDU list forms a VDU deployment command list.
  • An example of the VDU deployment command list that can be generated in this embodiment may be an example in the step S1120, and details are not described herein. A32.
  • the processor 55 may use the VM resource and the link resource determined by the step B to the VIM. Requesting to allocate corresponding resources, and loading software and performing basic data configuration for each VDU in the VDU list.
  • the performing the basic data configuration means that the software loaded for each VDU, the allocation result of the virtual machine VM resources allocated by each VDU, and the type of each VDU are recorded in the network element management device, and Each VDU is associated with an identifier of a VNF instance including the VDUs as a subsequent available resource configuration information, so that the network element management device manages the VNFs in the VNF instance and the instance based on the basic data configuration. It can be seen that the embodiment of the present invention automatically completes the basic data configuration in the process of performing the VNF instantiation, and solves the technical problem that the prior art needs to manually perform basic data configuration in the network element management device.
  • the processor 55 may execute the deployment command of each VDU in the command list one by one, and allocate each of the VDUs to the VIM request during the execution of each VDU deployment command.
  • the virtual machine VM resource is required, and according to the resource allocated by the VIM for each VDU, the software is loaded for the one VDU, and the allocation result of the required virtual machine VM resource is allocated to each VDU according to the VIM.
  • Each of the VDUs loads software and performs basic data configuration.
  • the processor 55 may request the VM resources and link resources required by all the VDUs in the VDU list to the VIM at one time, and load the software and perform basic data configuration for each VDU at one time after the request is completed. .
  • A33 After the deployment command of all VDUs in the VDU list is executed, the network element management device notifies the VNFM that the instantiated VNF is completed.
  • the processor 55 of the embodiment of the present invention performs the following steps before performing the step of generating a command list for deploying the VDU according to the determined VDU list:
  • An operation authorization request for instantiating a VNF is initiated to a Network Functions Virtualization Orchestrator (NFVO) via a VNFM (VNFM), and the instantiation is received from the NFVO via the VNFM a response of the operation authorization of the VNF, the response of the operation authorization of the instantiated VNF is that the NFVO determines that there is a resource required to execute the instantiated VNF and requests the VIM to reserve a corresponding resource, and then sends the response;
  • NFVO Network Functions Virtualization Orchestrator
  • VNFM VNFM
  • the processor 55 requests the VIM to allocate VM resources and link resources required for each VDU in the VDU list, and performs the following steps:
  • a request to allocate VM resources required for each VDU in the VDU list is forwarded to the VIM via the VNFM.
  • the processor 55 requests the VIM to allocate VM resources required for each VDU in the VDU list, and specifically performs the following steps:
  • a request to allocate VM resources required for each VDU in the VDU list is forwarded to the VIM via VNFM and NFVO in sequence.
  • the difference between this embodiment and the previous embodiment is that the NFVO directly initiates a resource allocation request to the VIM, and the previous embodiment further includes an operation authorization request and a response to determine that there are related resources in the network before initiating the resource allocation request.
  • the embodiment of the present invention transfers the lifecycle management work of the VNFM to the network element management device. Therefore, the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the network element management device does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to perform the lifecycle management of the VNF, and the corresponding actions are completed step by step, thereby avoiding the over-mechanization of the lifecycle management process of the instantiated VNF. Lack of flexibility.
  • the network element management device and the VNF usually belong to the same provider.
  • the embodiment of the present invention can support the difference in the performance of the VNF generated by each vendor, for example, the resource configuration information stored in the network element management device of different vendors.
  • the format may be different, and the specific processes of the network element management device of different vendors to instantiate the VNF according to the determined resources may also be different.
  • FIG. 35 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention. As shown in FIG. 35, it may include: an input device 60, an output device 61, a communication link 62, a transceiver device 63, a memory 64, and a processor 65, wherein:
  • the input device 60 is configured to receive input data externally input to the network element management device.
  • the input device 60 may include a keyboard, a mouse, a photoelectric input device, a sound input device, a touch input device, a scanner, and the like.
  • the input data can include various user instructions.
  • the output device 61 is configured to output output data of the network element management device.
  • the output device 61 described in the embodiment of the present invention may include a display, a speaker, a printer, and the like.
  • the output data can include response data to user instructions.
  • the communication link 62 is configured to communicate with other devices.
  • the communication link 62 described in the embodiment of the present invention may be an example of a propagation medium.
  • the propagation medium can generally embody computer readable instructions, data structures, program modules, or other data in the form of other modulated data signals, such as a carrier wave or other transport mechanism.
  • the communication medium can include wired media, such as a priority network or In a straight line connection, the propagation medium may also include a wireless medium such as sound waves, radio frequency, infrared rays, and the like.
  • the communication link 62 of the embodiment of the present invention may be used to transmit the performance parameter reported by the VNF, and the message for transmitting the VNF expansion completed to the VNFM, or for transmitting the VNF expansion request message.
  • the transceiver device 63 is configured to send data to or receive data from other devices through the communication link 62.
  • the transceiver device 63 of the embodiment of the present invention may be configured to receive the performance parameter reported by the VNF, the VNF instance expansion request message, and the message that the VNF instance is expanded and sent to the VNFM.
  • the transceiver device 63 can be a Ve-Vnfm interface with the VNFM.
  • the memory 64 is configured to store program data with various functions.
  • the data stored in the memory 64 in the embodiment of the present invention includes resource configuration information, and program data called and executed by the processor.
  • the memory 64 of an embodiment of the present invention may be a system memory such as volatile (such as RAM), non-volatile (such as ROM, flash memory, etc.), or a combination of both.
  • the memory 64 of the embodiment of the present invention may also be an external memory outside the system, such as a magnetic disk, an optical disk, a magnetic tape, or the like.
  • the processor 65 is configured to invoke program data and resource configuration information stored in the memory 64, and perform the following operations:
  • the performance parameter reported by the VNF obtained by the embodiment of the present invention may include a CPU usage rate, or any other performance parameter that can be reported by the VNF.
  • B2 determining, according to the reported performance parameter and the resource configuration information stored in the network element management device, a VDU list required for performing VNF instance expansion and a virtual machine VM resource required for each VDU in the VDU list; Or determining, according to the received VNF instance expansion request and the resource configuration information stored by the network element management device, a VDU list required to perform VNF instance expansion and a virtual machine VM resource required for each VDU in the VDU list.
  • B3 Perform VNF instance expansion according to the determined VDU list and virtual machine VM resources required by each VDU in the VDU list.
  • the resource configuration information stored in the memory 64 of the embodiment of the present invention may maintain a VNF instance expansion and contraction policy information, or the VNF instance expansion strategy information and a VNF may be separately maintained in the resource configuration information.
  • the example of the VNF instance expansion policy may include: when the performance parameter of the last report is lower than the second threshold, and the performance parameter of the current report is greater than or equal to the second threshold, performing VNF instance expansion, that is, On the basis of the existing VNF instance capacity specification, one or more specified types of VDUs are deployed.
  • the VNF instance shrinkage policy may include: when the performance parameter of the last report is greater than the first threshold, the performance parameter reported this time is smaller than When the difference between the first threshold and a buffer parameter is performed, the VNF instance is compacted to terminate one or more VDUs of the specified type in the VNF instance.
  • the description of the VNF expansion and contraction policy and the performance parameter of the VNF stored in this embodiment may be referred to the related description of step S211, and details are not described herein.
  • the processor 65 may determine the VDU list to be deployed for performing the VNF instance expansion, for example, when the foregoing policy is met. At 1 o'clock, it is determined on the basis of the original instance of the VNF that a VDU of type A needs to be deployed.
  • the resource configuration information stored in the memory 64 in the embodiment of the present invention may further include a ratio of the capacity specification to the virtual deployment unit VDU, a correspondence between the VDU and the software, a correspondence between the VDU and the virtual machine, or a network. Information such as interworking relationships and bandwidth requirements of each link. Therefore, in the embodiment of the present invention, after determining the VDU list to be deployed, the processor 65 may further determine the virtual machine VM resources required by each VDU in the VDU list according to the resource configuration information stored in the network element management device.
  • the virtual machine VM resource may include: a VM VM specification required by each VDU, a requirement of each VM to a link, and the like.
  • the resources determined by the embodiments of the present invention may further include: software corresponding to each VDU and affinity and isolation of each VM.
  • the processor 65 performs a VDU list and a VDU list that are required to be deployed in the VNF instance expansion according to the reported performance parameter and the resource configuration information stored in the network element management device.
  • steps of the VM VM resources required by the VDU may be performed:
  • VDU list Determine, according to the reported performance parameter and the VNF instance expansion policy information in the resource configuration information stored in the network element management device, a VDU list that needs to be deployed to perform VNF instance expansion. Specifically, how to determine the VDU list refers to the previous description, and details are not described herein.
  • B22 Determine, according to the correspondence between the VDU and the software in the resource configuration information, software corresponding to each VDU in the VDU list that is required to be deployed in the VNF instance expansion.
  • B23 Determine, according to the correspondence between the VDU and the virtual machine (VM) in the resource configuration information, the VM specifications required by each VDU in the VDU list that needs to be deployed in the VNF instance expansion.
  • the VM specifications required for different types of VDUs may be different.
  • An example of the correspondence between the VDU and the VM specification can be seen in the description of the step S2112, and details are not described herein.
  • B24 According to the network interworking relationship in the resource configuration information and the bandwidth requirement of each link, the VMs respectively require a link.
  • the example of the interworking relationship and the bandwidth requirement of the network element in the resource configuration information stored in the memory 64 may be as shown in step S2113, and details are not described herein.
  • the processor 65 determines the affinity and isolation of the VM according to the VDU isolation and affinity in the resource configuration information.
  • the example of the VDU isolation and the affinity in the resource configuration information stored in the memory 64 may be as shown in step S2114, and details are not described herein.
  • the processor 65 can determine the affinity and isolation of each VM according to the VDU isolation and affinity stored by the network element management device.
  • VM_A and VM_C are deployed on the same physical server, and multiple VM_A are deployed in different physical environments.
  • the types of VM_1 and VM_2 are VM_A, but when deployed, VM_1 and VM_2 need to be deployed on different physical servers.
  • the processor 65 is configured to determine, according to the received VNF instance expansion request and the resource configuration information stored by the network element management device, a VDU list that needs to be deployed to perform VNF instance expansion.
  • the virtual machine VM resources required by each VDU in the VDU list are as follows:
  • B26 Determine, according to the correspondence between the VDU and the software in the resource configuration information, software corresponding to each VDU in the VDU list that is required to be deployed in the VNF instance expansion;
  • B27 Determine, according to the correspondence between the VDU and the virtual machine (VM) in the resource configuration information, a VM specification required for each VDU in the VDU list to be deployed in the VNF instance expansion.
  • VM virtual machine
  • the VMs respectively require a link according to the network interworking relationship in the resource configuration information and the bandwidth requirement of each link.
  • the processor 65 when the processor 65 performs the step of performing VNF instance expansion according to the determined VDU list and the VM VM resources required for each VDU in the VDU list, the following steps are specifically performed:
  • VDU list includes one VDU of type A (simplified to: 1*VDU_A).
  • the memory 64 of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a deletion command RMV VDU, and the like. Therefore, the processor 65 can generate a deployment command for each VDU in the determined VDU list, and the deployment command set generated by each VDU in the entire VDU list forms a VDU deployment command list.
  • the determined VDU list is: 1*VDU_A
  • the embodiment of the present invention may generate the following VDU deployment command list:
  • VDU_1 (representing the number of the deployed VDU), VDU_A (representing the deployed VDU type);
  • the performing the basic data configuration means that the software loaded for each VDU, the allocation result of the virtual machine VM resources allocated by each VDU, and the type of each VDU are recorded in the network element management device, and Each VDU is associated with an identifier of a VNF instance including the VDUs as a subsequent available resource configuration information, so that the network element management device configures the VNF based on the basic data.
  • the VDUs in the examples and examples are managed. It can be seen that the basic data configuration is automatically completed in the process of performing the VNF instance expansion in the embodiment of the present invention, which solves the technical problem that the prior art needs to manually perform basic data configuration in the network element management device.
  • the processor 65 may execute the deployment command of each VDU in the command list one by one, and in the process of executing the deployment command of each VDU, request the VIM to be allocated for each VDU.
  • a virtual machine VM resource and according to the resource allocated by the VIM for each VDU, loading software for the one VDU and allocating a required virtual machine VM resource allocation result to each VDU according to the VIM,
  • Each VDU loads the software and performs basic data configuration.
  • the processor 65 may request the VM resources and link resources required by all the VDUs in the VDU list to the VIM at one time, and load the software and perform basic data configuration for each VDU once after the request is completed. .
  • the VNFM is notified that the VNF instance expansion is completed.
  • the processor 65 before performing the step of generating a command list for deploying a VDU according to the determined VDU list, the processor 65 further performs the following steps:
  • An operation authorization request for instantiating a VNF is initiated to a Network Functions Virtualization Orchestrator (NFVO) via a VNFM (VNFM), and the instantiation is received from the NFVO via the VNFM a response of the operation authorization of the VNF, the response of the operation authorization of the instantiated VNF is that the NFVO determines that there is a resource required to execute the instantiated VNF and requests the VIM to reserve a corresponding resource, and then sends the response;
  • NFVO Network Functions Virtualization Orchestrator
  • VNFM VNFM
  • the processor 65 requests the VMIM to allocate the VM resources required for each VDU in the VDU list
  • the VM 65 is specifically configured to allocate the VM resources required by each VDU in the VDU list via the VNFM. The request is forwarded to the VIM.
  • the processor 65 when the processor 65 requests the VMIM to allocate the VM resources required for each VDU in the VDU list, the processor 65 specifically allocates the VDUs in the VDU list through the VNFM and the NFVO. A request for the required VM resource is forwarded to the VIM.
  • the NFVO directly initiates a resource allocation request to the VIM, and the previous one
  • the embodiment further includes an operation authorization request and a response to determine that there are related resources in the network before initiating the resource allocation request.
  • the embodiment of the present invention transfers the lifecycle management work of the VNFM to the network element management device. Therefore, the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the lifecycle management of the VNF instance expansion does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to complete the corresponding actions step by step. This avoids the over-mechanization of the lifecycle management process of the VNF instance expansion. Lack of flexibility.
  • the network element management device and the VNF are generally the same vendor. Therefore, the embodiment of the present invention can support the difference in the VNF instance expansion of each vendor, for example, the resource configuration information stored in the network element management device of different vendors.
  • the format of the VNF instance expansion may be different according to the determined resources.
  • FIG. 36 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention. As shown in FIG. 36, it may include: an input device 70, an output device 71, a communication link 72, a transceiver device 73, a memory 74, and a processor 75, wherein:
  • the input device 70 is configured to receive input data externally input to the network element management device.
  • the input device 70 may include a keyboard, a mouse, a photoelectric input device, a sound input device, a touch input device, a scanner, and the like.
  • the input data can include various user instructions.
  • the output device 71 is configured to output output data of the network element management device.
  • the output device 71 may include a display, a speaker, a printer, and the like.
  • the output data can include response data to user instructions.
  • the communication link 72 is configured to communicate with other devices.
  • the communication link 72 described in the embodiment of the present invention may be an example of a propagation medium.
  • the propagation medium can generally embody computer readable instructions, data structures, program modules, or other data in the form of other modulated data signals, such as a carrier wave or other transport mechanism.
  • the communication medium can include wired media, such as a priority network or In a straight line connection, the propagation medium may also include a wireless medium such as sound waves, radio frequency, infrared rays, and the like.
  • the communication link 72 of the embodiment of the present invention can be used to transmit performance parameters and transmissions reported by the VNF. The message that the VNF is sent to the VNFM to complete the shrinkage or is used to transmit the VNF contraction request message.
  • the transceiver device 73 is configured to send data to or receive data from other devices through the communication link 72.
  • the transceiver device 73 of the embodiment of the present invention may be configured to receive the performance parameter reported by the VNF, the VNF instance shrink request message, and the message that the VNF instance is compressed and sent to the VNFM.
  • the transceiver device 73 can be a Ve-Vnfm interface with the VNFM.
  • the memory 74 is configured to store program data with various functions.
  • the data stored in the memory 74 in the embodiment of the present invention includes resource configuration information, and program data called and executed by the processor.
  • the memory 74 of an embodiment of the present invention may be a system memory such as volatile (such as RAM), non-volatile (such as ROM, flash memory, etc.), or a combination of both.
  • the memory 74 of the embodiment of the present invention may also be an external memory outside the system, such as a magnetic disk, an optical disk, a magnetic tape, or the like.
  • the processor 75 is configured to invoke program data and resource configuration information stored in the memory 74, and perform the following operations:
  • C1 Obtain performance parameters reported by the VNF or receive a VNF instance shrink request
  • the performance parameter reported by the VNF obtained by the embodiment of the present invention may include a CPU usage rate, or any other performance parameter that can be reported by the VNF.
  • the VDU list to be terminated includes one or more VDUs to be terminated
  • the VNF instance shrinking policy in the resource configuration information includes: when the performance parameter of the last report is greater than the first threshold, the performance parameter reported this time When the difference between the first threshold and the buffer parameter is smaller, performing VNF instance shrinking, terminating one or more specified types of VDUs; or, according to the received VNF instance shrink request and the VNF in the resource configuration information
  • the instance shrinking policy determines a VDU list to be terminated by performing VNF instance shrinkage and a VM resource to be released corresponding to each VDU in the VDU list.
  • C3 Perform VNF instance reduction according to the determined VDU list to be terminated and the VM resources to be released corresponding to each VDU in the VDU list.
  • the embodiment of the present invention can be maintained in the resource configuration information stored in the memory 74.
  • a VNF instance expansion policy information, or a VNF instance expansion policy information and a VNF instance reduction policy information may be separately maintained in the resource configuration information, where the VNF instance expansion policy may include:
  • the reported performance parameter is lower than the second threshold.
  • the performance parameter of the current report is greater than or equal to the second threshold, the VNF instance is expanded, that is, one or more specified types are deployed based on the existing VNF instance capacity specifications.
  • the VNF instance shrinkage policy may include: when the performance parameter of the last report is greater than the first threshold, and the performance parameter of the current report is smaller than the difference between the first threshold and a buffer parameter, performing VNF instance shrinkage and terminating One or more specified types of VDUs in a VNF instance.
  • the description of the VNF expansion and contraction policy and the performance parameter of the VNF stored in this embodiment may be referred to the related description of step S311, and details are not described herein.
  • the processor 75 may determine the VDU list to be terminated by performing the VNF instance shrinking, for example, when the above is satisfied.
  • strategy 3 it is determined that a VDU of type A needs to be terminated on the basis of the original instance of the VNF.
  • the resource configuration information stored in the memory 74 in the embodiment of the present invention may further include a ratio of the capacity specification to the virtual deployment unit VDU, a correspondence between the VDU and the software, a correspondence between the VDU and the virtual machine, and network interworking. Information such as relationships and bandwidth requirements for each link. Therefore, in the embodiment of the present invention, after determining the VDU list to be terminated, the processor 75 may further determine, according to the resource configuration information stored in the network element management device, the virtual machine VM resources to be released by each VDU in the VDU list. And the software that each VDU needs to delete.
  • the virtual machine VM resource may include: a VM VM specification required by each VDU, a requirement of each VM to a link, and the like.
  • the processor 75 of the embodiment of the present invention may further determine a VDU having an affinity and isolation relationship with the determined VDU to be terminated, which is to be terminated.
  • the processor 75 of the embodiment of the present invention determines, according to the reported performance parameter and the VNF instance shrinkage policy in the resource configuration information, a VDU list and a VDU that are to be terminated by performing VNF instance shrinkage.
  • the VM resources to be released corresponding to the VDUs in the list are as follows:
  • the VNF instance in the shrinking policy information determines the list of VDUs that need to be terminated to perform the VNF instance shrinking. Specifically, how to determine the previous description is not described here.
  • C22 Determine, according to the basic data configuration information of each VDU in the resource configuration information, software that needs to be deleted in each VDU in the VDU list to be terminated, a VM specification to be released, and a link related to the VM to be released.
  • VM specifications corresponding to different types of VDUs may be different.
  • the example of the interworking relationship and the bandwidth requirement of the network element in the resource configuration information stored in the network element management device in this embodiment may be as shown in step S3111, and details are not described herein. Therefore, the processor 75 of the embodiment of the present invention can determine that the link between VM_A and VM_B needs to be released according to the foregoing interworking relationship, and release the link between VM_B and VM_C.
  • the processor 65 may also delete the basic data of the VDU configured in the network element management unit. For example, the association relationship between the VDU and the VNF instance identifier is deleted, the stored correspondence relationship between the VDU and the software is deleted, the correspondence between the VDU and the VM specification is deleted, and the related chain of the VM of the VDU is deleted. Road information, etc.
  • the processor 75 may further determine, according to the VM isolation and affinity in the resource configuration information, a VDU that has an affinity and isolation relationship with the VDU in the VDU list to be terminated.
  • the example of the VM isolation and the affinity in the resource configuration information stored in the memory 74 of this embodiment may be as shown in step S31112, and details are not described herein. Therefore, for example, it is determined that the VDU_A needs to be terminated, and the physical server is terminated.
  • the VDU of the type VDU_1 of the VDU_A is 1 on the VDU
  • the VDU of the VDU_3 of the type VDU_C located on the physical server 1 is also terminated.
  • the processor 75 in the embodiment of the present invention performs the VNF instance shrinkage request according to the received VNF instance shrink request and the resource configuration information, and determines that the VNF instance shrinkage is to be terminated. a VDU list and a VM to be released corresponding to each VDU in the VDU list When the steps of the resource are performed, the following steps can be performed:
  • C24 Determine, according to the received contention VNF instance request and the resource configuration information stored by the network element management device, a VDU list that is required to be terminated by performing VNF instance shrinkage, and a corresponding release of each VDU in the VDU list.
  • VM resources
  • the processor 75 determines, according to the received contention VNF instance request and the resource configuration information stored by the network element management device, the VDU list and the VDU that are required to terminate the VNF instance shrinkage.
  • the VM resources to be released corresponding to each VDU in the list are as follows:
  • the C242 determines, according to the basic data configuration information of each VDU in the resource configuration information, the software to be deleted in each VDU in the VDU list to be terminated, the VM specification to be released, and the link related to the VM to be released.
  • the processor 75 may further determine, according to the VM isolation and affinity in the resource configuration information, a VDU that has an affinity and isolation relationship with the VDU in the VDU list to be terminated.
  • the processor 75 performs the following steps according to the step of performing the VNF instance shrinking according to the determined VDU list to be terminated and the VM resources to be released corresponding to the VDUs in the VDU list:
  • the network element management device of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a deletion command RMV VDU, and the like. Therefore, the processor 75 can generate a delete command for each VDU in the determined VDU list, and the delete command set generated by each VDU in the entire VDU list forms a VDU delete command list.
  • the determined VDU list is: 1*VDU_A
  • the embodiment of the present invention may generate the following VDU deletion command list:
  • VDU_1 (representing the number of the terminated VDU), VDU_A (representing the type of the terminated VDU);
  • the embodiment of the present invention transfers the lifecycle management work of the VNFM to the network element management device. Therefore, the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the lifecycle management of the VNF instance shrinkage does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to complete the corresponding actions step by step, avoiding the lifecycle management process of the VNF instance shrinkage. Mechanization and lack of flexibility.
  • the network element management device and the VNF usually belong to the same vendor. Therefore, the embodiment of the present invention can support the difference in the VNF instance shrinkage of each vendor, for example, the resource configuration stored in the network element management device of different vendors.
  • the format of the information may be different, and the specific processes of the network element management device of different vendors to perform VNF instance shrinkage according to the determined resources may also be different.
  • the ping-pong elastic expansion of the VNF is reduced by a buffer parameter.
  • FIG. 37 is a schematic structural diagram of another embodiment of a network element management device according to an embodiment of the present invention. As shown in FIG. 37, it may include an input device 80, an output device 81, a communication link 82, a transceiver device 83, a memory 84, and a processor 85, wherein:
  • the input device 80 is configured to receive input data externally input to the network element management device.
  • the input device 80 may include a keyboard, a mouse, a photoelectric input device, a sound input device, a touch input device, a scanner, and the like.
  • the input data can include various user instructions.
  • the output device 81 is configured to output output data of the network element management device.
  • the output device 81 may include a display, a speaker, a printer, and the like.
  • the output data can include response data to user instructions.
  • the communication link 82 is configured to communicate with other devices.
  • the communication link 82 described in the embodiment of the present invention may be an example of a propagation medium.
  • the propagation medium can generally embody computer readable instructions, data structures, program modules, or other data in the form of other modulated data signals, such as a carrier wave or other transport mechanism.
  • the communication medium can include wired media, such as a priority network or In a straight line connection, the propagation medium may also include a wireless medium such as sound waves, radio frequency, infrared rays, and the like.
  • the communication link 82 of the embodiment of the present invention may be used to transmit a message that terminates the VNF instance request message and terminates the completion of the VNF instance sent to the VNFM.
  • the transceiver device 83 is configured to send data to or receive data from other devices through the communication link 82.
  • the transceiver device 83 of the embodiment of the present invention may be configured to receive a message that terminates the VNF instance request message and sends a VNF instance termination completion to the VNFM.
  • the transceiver device 83 can be a Ve-Vnfm interface with the VNFM.
  • the memory 84 is configured to store program data with various functions.
  • the data stored by the memory 84 in the embodiment of the present invention includes resource configuration information, and program data called and executed by the processor.
  • the memory 84 of an embodiment of the present invention may be a system memory such as volatile (such as RAM), non-volatile (such as ROM, flash memory, etc.), or a combination of both.
  • the memory 84 of the embodiment of the present invention may also be an external memory outside the system, such as a magnetic disk, an optical disk, a magnetic tape, or the like.
  • the processor 85 is configured to invoke program data and resource configuration information stored in the memory 84, and perform the following operations:
  • D1 Receive the terminating VNF instance request message sent by the VNF manager VMFM.
  • D2 Determine, according to the terminating VNF instance request message and the resource configuration information stored in the network element management device, a VDU list that needs to be terminated to terminate the VNF instance, and a VM resource to be released corresponding to each VDU in the VDU list.
  • D3 Performing the VNF instance termination according to the determined VDU list to be terminated and the VM resources to be released corresponding to each VDU in the VDU list.
  • the processor 85 in the embodiment of the present invention performs execution of determining to terminate the VNF instance according to the terminating VNF instance request message and the resource configuration information stored in the network element management device.
  • the steps of the VDU to be terminated and the VM resources to be released corresponding to the VDUs in the VDU list are performed, the following steps are performed:
  • D21 Determine, according to the VNF instance identifier carried in the termination VNF instance request message and the resource configuration information stored in the network element management device, the VNF instance to be terminated and the VDU list to be terminated;
  • the basic data configuration information of each VDU in the resource configuration information determines software to be deleted for each VDU in the VDU list to be terminated, a VM specification to be released, and a link related to the VM to be released.
  • VM specifications corresponding to different types of VDUs may be different.
  • an example of the correspondence between the VDU and the VM specification in this embodiment may be an example in the step S4111, and details are not described herein.
  • the processor 85 of the embodiment of the present invention can determine the VM to be deleted according to the correspondence between the VDU and the VM specification. For example, when it is determined that the VDU list needs to be terminated as 1*VDU_A+1*VDU_B+1*VDU_C, it may be determined that one VM_A needs to be deleted, and the VM_A is deleted to release 4 cores vCPU+8G memory+2G bandwidth; VM_B, delete the VM_B specifically to release 4 core vCPU + 16G memory + 2G bandwidth; need to delete 1 VM_C, delete the VM_C specifically to release 2 core vCPU + 8G memory + 2G bandwidth.
  • the example of the interworking relationship and the bandwidth requirement of the network element in the resource configuration information stored in the memory 84 of the embodiment may be as shown in step S4111, and details are not described herein.
  • the processor 85 of the embodiment of the present invention can determine that the link between VM_A and VM_B needs to be released according to the foregoing interworking relationship, and release the link between VM_B and VM_C.
  • the processor 85 may further determine, according to the VM isolation and affinity in the resource configuration information, a VDU that has an affinity and isolation relationship with the VDU in the VDU list to be terminated.
  • VM isolation and affinity in the resource configuration information stored in the memory 84 of the embodiment may be shown in step S41112, and details are not described herein.
  • VDU_A if it is determined that the VDU_A needs to be terminated, when the VDU of the type VDU_1 of the type VDU_A located on the physical server 1 is terminated, the type on the physical server 1 is also terminated.
  • the VDU_C is numbered VDU of VDU_3.
  • the processor 85 of the embodiment of the present invention performs the steps of terminating the VNF instance according to the determined VDU list to be terminated and the VM resources to be released corresponding to the VDUs in the VDU list, and the specific execution is as follows: step:
  • the memory 84 of the embodiment of the present invention stores various types of configuration commands, such as a deployment command ADD VDU, a deletion command RMV VDU, and the like. Therefore, the processor 85 can generate a delete command for each VDU in the determined VDU list, and the delete command set generated by each VDU in the entire VDU list forms a VDU delete command list.
  • an example of the VDU deletion command list generated in this embodiment may be an example in the step S4121, and details are not described herein.
  • D32 in the process of executing the VDU delete command in the command list for deleting the VDU, uninstalling the software of each VDU, and requesting the VIM to release the virtual machine VM resources related to each VDU through the VNFM, and the network element is The underlying data associated with the VDU in the management device is deleted.
  • the embodiment of the present invention transfers the lifecycle management work of the VNFM to the network element management device. Therefore, the VNFD does not need to set a complete parameter or script that can cover each VNF lifecycle management process.
  • the network element management device performs the lifecycle management of the VNF instance termination, it does not need to be instructed by the corresponding parameters or processes and scripts in the VNFD to complete the corresponding actions step by step, thereby avoiding the over-mechanization of the lifecycle management process of the VNF instance termination. Lack of flexibility.
  • the network element management device and the VNF are generally in the same vendor.
  • the embodiment of the present invention can support the difference in the termination of the VNF instance by each vendor, for example, the resource configuration information stored in the network element management device of different vendors.
  • the format of the VNF instance termination may be different, and the network management device of different vendors may perform the VNF instance termination according to the determined resources.
  • the embodiment of the VNF lifecycle management method and the network element management device of the embodiment of the present invention is The detailed flow of the VNF instantiation process, the VNF expansion/reduction process, and the VNF instantiation termination process of the embodiment of the present invention will be described below in conjunction with a specific application scenario.
  • FIG. 38 is a schematic flow chart of the VNF instantiation process of the present invention. As shown in FIG. 38, it may include:
  • the VNFM receives the instantiation VNF request, the requesting party can be the NFVO, etc.; the VNF information model carried in the request describes the template VNFD index information and the capacity specification, and queries the corresponding VNFD to query the type of the VNF to be deployed, the version number and Capacity specification. Alternatively, the type, version number, and capacity specification of the VNF to be carried are determined to determine the type, version number, and capacity specification of the VNF to be deployed.
  • the VNFM determines the EMS (the network element management device used in this example) according to the information such as the Vendor/VNF type, and forwards the request to the EMS.
  • the EMS determines, according to the information carried in the request and the stored resource configuration information, the resources required to perform the instantiation of the VNF.
  • the EMS carries the resources required to perform the instantiation of the VNF, and initiates an authorization request for instantiating the VNF operation to the VNFM.
  • the VNFM forwards the authorization request for instantiating the VNF operation to the NFVO.
  • NFVO checks the resource situation in the resource pool, can meet the requirements of the instantiation; if not, the authorization fails; otherwise, S07 is executed;
  • the NFVO initiates a resource reservation request to the VIM.
  • VIM reserves corresponding resources for this instantiation requirement, and responds to NFVO (with VIM identifier).
  • VFVO returns an instantiated VNF operation authorization response to the VNFM.
  • VNFM forwards this response to the EMS.
  • the EMS converts the VDU list determined by S03 into a corresponding command list for deploying the VDU.
  • the command list of the deployed VDU generated by the EMS is as follows:
  • S11 The EMS executes each command (such as the first command), internally generates a requirement for the VM specification and the link, and initiates a resource allocation request to the VNFM.
  • the VNFM initiates a resource allocation request to the VIM determined by S08.
  • the VIM allocates the VM_A type VM_1 for the request, and establishes a virtual network connection with the neighboring VM for the VM_1 according to the link requirement in the request, and the allocation successfully returns the VM index (VM_1).
  • the VNFM forwards the resource allocation response to the EMS.
  • the EMS deploys the software of the VDU_A to complete some basic data configuration, and records the basic data configuration information of the VDU_1 as the resource configuration information in the data tables of the EMS and the VNF, so that the EMS can subsequently manage the VDU_1.
  • the EM loop executes each configuration command (ie, cyclic execution S11-S16) to complete the deployment of each required VDU, and after all the VDU deployments are completed, notify the VNFM that the instantiation is completed.
  • each configuration command ie, cyclic execution S11-S16
  • VNFM informs NFVO that the instantiation is completed, and NFVO refreshes the resource usage information of this instance.
  • the VNF instance expansion process is similar to the VNF instantiation process. The difference is that the number of VDUs to be processed is different and the policy for determining whether to expand capacity is different.
  • the specific capacity expansion policy has been introduced in the foregoing embodiments. Therefore, the scenario description of the VNF instance expansion process is not performed here.
  • FIG. 39 is a schematic flow chart showing the process of terminating a VNF instance according to the present invention. As shown in FIG. 39, it may include:
  • the VNFM receives the request to terminate the VNF instance, and the requesting party is NFVO;
  • the VNFM forwards the request to the corresponding EMS.
  • the EMS determines, according to the stored resource configuration information, the VNF to be terminated and the VDU list that is included, and generates a command list for deleting the VDU, where the list includes a deletion command of the VDU to be deleted corresponding to the VNF instance.
  • the EMS initiates a request to uninstall a VDU to the VNF, and the VNF uninstalls the software corresponding to the VDU.
  • the EMS executes the VDU delete command to delete the basic data related to the VDU on the EMS.
  • the EMS initiates a resource release request to the VNFM, requesting to release the VM related to the VDU. And links.
  • S24 and S25 are in no particular order, but it is necessary to ensure that the VM release identifier can be carried in the resource release request, so that the VIM can be deleted correctly.
  • the VNFM forwards the request to the corresponding VIM.
  • the VIM releases the VM and associated link resources and replies with a resource release response.
  • VNFM forwards this response to the EMS.
  • the EMS cyclically executes S23-S29, and releases all VDUs and their resources in the VNF to be terminated. If there is still this VNF related data on the EMS, it is also deleted.
  • the EMS returns a VNF instance termination completion response to the VNFM.
  • the VNFM deletes the data related to the VNF instance, and returns a VNF instance termination completion response to the NFVO, and the NFVO also deletes the data information related to the VNF instance.
  • the process of shrinking the VNF instance is similar to the process of terminating the VNF instance.
  • the difference is that the number of processed VDUs is different and the strategy for determining whether to shrink is different.
  • the specific shrinkage strategy has been introduced in the previous embodiment. . Therefore, here, the scene description of the VNF instance shrinking process is not performed.
  • an embodiment of the present invention further provides a computer storage medium, where the computer storage medium can store a program, and the program can execute some or all of the steps of the method according to the embodiment of the present invention.
  • the computer storage medium of the embodiment of the present invention includes: RAM, ROM, EEPROM, flash memory, CD-ROM, DVD or other optical storage, magnetic tape, magnetic disk or other magnetic storage, or any other information that can be used for storing information.
  • the embodiment of the present invention has the following advantages over the prior art VNFM (which can be the VNFM in the architecture shown in FIG. 1), which relies on the VNFD to perform the corresponding VNF lifecycle management:
  • VNFD provided by each Vendor is no longer required to cover all lifecycle management processes, nor does VNFD need to be standardized.
  • Vendor management devices of different Vendors can reflect the differences of Vendors.
  • the network element management device receives a lifecycle management operation, it is no longer necessary to perform the corresponding actions step by step according to the corresponding parameters of the VNFD or the instructions of the script and the script, so that the life cycle management process of the VNF is performed. Very flexible.
  • the final VNFD can be simplified as follows:
  • Simplified mode 1 Only the vendor (corresponding to the parameter of the parameter column in Table 1 and its corresponding parameter value), the VNF type (corresponding to the VNF id of the parameter column in Table 1 and its corresponding parameter value), and the version (correspondence table) are reserved in the VNFD.
  • Basic information such as the VNF version of the parameter column in 1 and its corresponding parameter value), the capacity specification (corresponding to the capacity of the parameter column in Table 1 and its corresponding parameter value).
  • Other information in the prior art VNFD does not need to be retained.
  • Simplified mode 2 The entire VNFD is removed, but the NSD template needs to be modified to ensure that the Vendor, VNF type, version information, and capacity specifications that can be obtained through the associated VNFD can be reflected in the NSD.
  • the NFVO->VNFM interface includes the following information as long as the method of the embodiment of the present invention is executed:
  • Vendor+VNF type used by the VNFM to forward the request to the correct NE management device
  • the VNF type, the VNF version number, and the capacity specification are forwarded by the VNFM to the network element management device, so that the network element management device uses this decision to instantiate which version of the VNF, and deploys according to the size.
  • the embodiment of the present invention needs to retain the parameters and scripts of the LMU in the VNFD or the NSD, so that the embodiment of the present invention performs VNF lifecycle management through the LMU.
  • the LMU could be deployed first.
  • the deployment process of the LMU is the same as that in the VNFD deployment process in the prior art, and is not described here.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明实施例公开了一种实例化VNF的方法及网元管理设备。其中，所述方法包括：网元管理设备接收VNF管理器VNFM发送的实例化VNF请求消息；所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源；所述网元管理设备根据所述确定出的资源来实例化VNF。实施本发明实施例之后，不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了实例化VNF的生命周期管理过程过于机械化和缺乏灵活性。

Description

一种实例化VNF的方法及网元管理设备
本申请要求了2014年11月6日提交的、申请号为201410621561.3、发明名称为“一种实例化VNF的方法及网元管理设备”的中国申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本发明涉及通信领域，尤其涉及一种实例化VNF的方法及网元管理设备。
背景技术
2012年13家Top运营商(AT&T、Verizon、VDF、DT、T-Mobile、BT、Telefornica等)发起成立ETSI NFV ISG，目标是基于现代化的虚拟化技术，网络设备可以基于通用硬件，如基于标准的大容量Server、存储和以太网，不同Vendor提供的应用以软件形式，远程自动部署在统一的云虚拟化基础设施上；从而降低成本，提高效率、增加敏捷能力。
在NFV ISG定义的网络功能虚拟化(Network Functions Virtualization，NFV)的参考框架如图1所示，包括：业务支持系统(Operation Support System，OSS)、数据仓库(service，VNF and Infrastructure Description)，网元管理系统(Element Management System，EMS)、虚拟化的网络功能(Virtual Network Function，VNF)、NFV的基础设施层(NFV Infrastructure，NFVI)、NFV编排器(NFV Orchestrator，NFVO)、VNF管理器(VNF Manager)、虚拟化基础设施管理系统(Virtual Infrastructure Manager，VIM)，其中：
OSS除传统的管理功能外，针对虚拟化应用与Orchestrator交互完成对网络的操作和维护。
数据仓库，包含相关的模板及VNF软件镜像等。
NFV Orchestrator主要功能是实现运营商域内(1或多个数据中心)的网络业务(Network Service，NS)生命周期管理(如部署/扩容/缩容/下线等)，以及NFVI资源编排、策略管理等功能；NFV Orchestrator根据NS的信息模型描述模版(NS  Descriptor，NSD)分解出对各VNF的需求，配合VNFM实现VNF的部署。
VNFM主要功能是实现VNF的生命周期管理，如部署/扩容/缩容/下线等自动化能力；VNFM根据VNF的信息模型描述信息(VNF Descriptor，VNFD)模板及VNF容量需求，分解出对虚拟机等虚拟资源的需求，与NFV Orchestrator、VIM配合完成VNF的实例化。
VIM主要功能是实现对整个基础设施层资源(含计算、存储、网络资源)的管理和监控。
目前，运营商向供应商(Vendor)购买VNF，Vendor提供按照NFV Orchestrator可以理解的格式完成VNFD参数的输入，以及相关的软件数据包，运营商将数据包注册到NFV Orchestrator的目录(Catalog)中。
当前NFV规范中对VNFD定义的非常详细(表1仅作参考)，详细内容参考NVF MAN001GS文档。
表1：
Figure PCTCN2015092869-appb-000001
Figure PCTCN2015092869-appb-000002
现有技术中，VNFM负责VNF的生命周期的管理过程，而每个过程各vendor厂家实现都有差异性，要做到统一管理，就必须要求各Vendor提供的VNFD要能覆盖各个过程，而且要能做到标准化，无法体现各vendor的差异性。并且，当VNFM收到一个生命周期(Lifecycle)管理的操作(触发者可以是NFVO、 VNFM自身、EMS等)，解析VNFD，然后根据VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，使得VNF的生命周期管理过程非常机械化，缺乏灵活性。
发明内容
本发明实施例提供一种实例化VNF的方法及网元管理设备，可简化VNF实例化VNF的流程，并能支持各供应商的网元管理设备进行VNF实例化的差异性管理。
本发明第一方面提供一种实例化虚拟化的网络功能VNF的方法，其可包括：
网元管理设备接收VNF管理器VNFM发送的实例化VNF请求消息；
所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源；
所述网元管理设备根据所述确定出的资源来实例化VNF。
结合第一方面，在第一种可行的实施方式中，
所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源具体包括：
所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定实例化VNF所需的虚拟部署单元VDU列表，以及所述VDU列表中各VDU所需虚拟机VM资源；
所述网元管理设备根据所述确定出的资源来实例化VNF具体包括：
所述网元管理设备根据所述确定的VDU列表生成部署VDU的命令列表；
所述网元管理设备在执行所述命令列表中的VDU的部署命令过程中，向虚拟化基础设施管理系统VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置；
所述网元管理设备在所述VDU列表中所有VDU的部署命令执行完之后，通知所述VNFM所述实例化VNF完成。
结合第一方面的第一种可行的实施方式，在第二种可行的实施方式中，
所述网元管理设备在执行所述命令列表中的VDU的部署命令过程中，向所述VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置具体包括：
所述网元管理设备逐一执行所述命令列表中的每一个VDU的部署命令，在执行每一个VDU的部署命令过程中，向所述VIM请求分配所述每一个VDU所需的虚拟机VM资源，并根据所述VIM为所述每一个VDU分配的资源为所述每一个VDU加载软件，并根据所述VIM对每一个VDU分配所需的虚拟机VM资源的分配结果，为所述每一个VDU加载软件并进行基础数据配置。
本发明第二方面提供一种网元管理设备，其可包括：
接收模块，用于接收VNF管理器VNFM发送的实例化VNF请求消息；
确定模块，用于根据所述接收模块接收的所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源；
实例化模块，用于根据所述确定模块所确定出的资源来实例化VNF。
由上可见，本发明实施例中网元管理设备接收VNF管理器VNFM发送的实例化VNF请求消息；并根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源；以及根据所述确定出的资源来实例化VNF具体实现中，网元管理设备与VNF属于同一供应商，不同的供应商之间的网元管理设备无需做到标准化和统一化，因此，本发明实施例能支持各供应商执行实例化VNF时的差异性，比如，不同供应商中存储的资源配置信息的格式就可以不相同，以及不同供应商中根据确定出的资源来实例化VNF的具体流程也可不同。另外，本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行实例化VNF的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相 应的动作，避免了实例化VNF的生命周期管理过程过于机械化和缺乏灵活性。
附图说明
图1为NFV ISG定义的网络功能虚拟化(Network Functions Virtualization，NFV)的参考框架；
图2为本发明实施例的实例化虚拟化的网络功能VNF的方法的流程示意图；
图3为图2所示的方法流程中步骤S111的一实施例流程示意图；
图4为图2所示的方法流程中步骤S112的一实施例流程示意图；
图5为本发明实施例的VNF实例扩容的方法的一实施例流程示意图；
图6为图5所示的方法流程中步骤S211的一实施例流程示意图；
图7为本发明实施例的VNF实例扩容的方法的另一实施例流程示意图；
图8为图7所示的方法流程中步骤S221的一实施例流程示意图；
图9为图5或图7所示的方法流程中步骤S212的一实施例流程示意图；
图10为本发明实施例的VNF实例缩容的方法的一实施例流程示意图；
图11为图10所示的方法流程中步骤S311的一实施例流程示意图；
图12为本发明实施例的VNF实例缩容的方法的另一实施例流程示意图；
图13为图12所示的方法流程中步骤S321的一实施例流程示意图；
图14为图10或图12所示的方法流程中步骤S312的一实施例流程示意图；
图15为本发明实施例的终止VNF实例的方法的流程示意图；
图16为图15所示的方法流程中步骤S411的一实施例流程示意图；
图17为图15所示的方法流程中步骤S412的一实施例流程示意图；
图18为本发明实施例的网元管理设备的一实施例的结构组成示意图；
图19为图18中确定模块11的一实施例的结构组成示意图；
图20为图18中实例化模块12的一实施例的结构组成示意图；
图21为本发明实施例的网元管理设备的另一实施例的结构组成示意图；
图22为图21中确定模块21的一实施例的结构组成示意图；
图23为图21中扩容模块22的一实施例的结构组成示意图；
图24为本发明实施例的网元管理设备的另一实施例的结构组成示意图；
图25为图24中确定模块24的一实施例的结构组成示意图；
图26为本发明实施例的网元管理设备的另一实施例的结构组成示意图；
图27为图26中确定模块31的一实施例的结构组成示意图；
图28为图26中缩容模块32的一实施例的结构组成示意图；
图29为本发明实施例的网元管理设备的另一实施例的结构组成示意图；
图30为图29中确定模块34的一实施例的结构组成示意图；
图31为本发明实施例的网元管理设备的另一实施例的结构组成示意图；
图32为图31中确定模块41的一实施例的结构组成示意图；
图33为图31中终止模块42的一实施例的结构组成示意图；
图34为本发明实施例的网元管理设备的另一实施例的结构组成示意图；
图35为本发明实施例的网元管理设备的另一实施例的结构组成示意图；
图36为本发明实施例的网元管理设备的另一实施例的结构组成示意图；
图37为本发明实施例的网元管理设备的另一实施例的结构组成示意图；
图38为本发明的VNF实例化过程的原理流程示意图；
图39为本发明的终止VNF实例过程的原理流程示意图。
具体实施方式
为使本发明的目的、技术方案和优点更加清楚，下面将结合附图对本发明作进一步地详细描述。
图2为本发明实施例的实例化虚拟化的网络功能VNF的方法的流程示意图。如图2所示，其可包括：
步骤S110，网元管理设备接收VNF管理器VNFM发送的实例化VNF请求消息。
具体实现中，本发明实施例的所述网元管理设备与所述VNF由同一供应商提供，可包括但不限于网元管理系统(Element Management System，EMS)或 VNF本地管理单元(Local Management Unit，LMU)。所述网元管理设备可通过与VNFM之间的Ve-Vnfm接口从所述VNFM接收实例化VNF请求消息。
步骤S111，所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源。具体实现中，所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定实例化VNF所需的虚拟部署单元VDU列表，以及所述VDU列表中各VDU所需虚拟机VM资源。其中，所述虚拟机VM资源可包括：各VDU所需的虚拟机VM规格、以及各VM对链路的需求等。进一步，作为补充实施例，本发明实施例所确定的执行实例化VNF所需的资源还可包括：各VDU对应的软件及各VM的亲和度和隔离度。
具体实现中，参考图3，作为一种实施例，步骤S111可包括如下步骤：
步骤S1110，所述网元管理设备根据所述实例化VNF的请求确定要部署的VNF的类型、版本号及容量规格。
具体实现中，在步骤S1110，所述网元管理设备可根据所述实例化VNF请求中携带的VNF的信息模型描述模版VNFD索引信息和容量规格，到对应的VNFD中查询要部署的VNF的类型，版本号及容量规格。例如，假设所述实例化VNF的请求中携带的VNFD索引为表1的VNFD的索引xx，容量规格为表1中的“部署规格1”，则在步骤S1110网元管理设备便可根据所述索引xx找到表1所述的VNFD，进而可在表1中查询到要部署的VNF的类型(对应表1中的“VNF id”)、版本号(对应表1中的“VNF版本”)及容量规格(对应表1中的“部署规格1”)。具体实现中，关于容量规格的携带可仅携带容量规格编号(对应表1中的部署规格1的编号1)，也可携带容量规格的容量大小(对应表1中的容量500kcaps)。
或者，在步骤S1110，所述网元管理设备可根据所述实例化VNF请求携带的VNF的类型、版本号及容量规格确定要部署的VNF的类型、版本号及容量规格。例如，所述实例化VNF的请求携带的VNF的类型为CSCF，版本号为1，容量规格为部署规格1(对应部署规格1所列表项)。具体实现中，关于容量规 格的携带可仅携带容量规格编号(比如，表1中的部署规格1的编号1)，也可携带容量规格的容量大小(比如，表1中的部署规格1的容量500kcaps)。由此，在步骤S1110可确定要部署的VNF的类型为CSCF，版本号为1，容量规格为500kcaps的容量。
具体实现中，所述实例化VNF请求携带的VNF的类型、版本号及容量规格注册在所述VNFD中或注册在网络业务的信息模型描述模版NSD中。
步骤S1111，所述网元管理设备根据所述资源配置信息中的容量规格与虚拟部署单元VDU的配比关系，确定所述要部署的VNF所需的VDU列表。
本发明实施例中，网元管理设备存储的资源配置信息中包括容量规格与虚拟部署单元VDU的配比关系(比如，表1中的VDU列表项中的“1*VDU_A+1*VDU_B+1*VDU_C”)，因此，在步骤S1111，网元管理设备便可根据存储的容量500kcaps与VDU列表1*VDU_A+1*VDU_B+1*VDU_C的对应关系，确定编号为1的容量规格或容量为500kcaps的容量规格对应的VDU列表为1*VDU_A+1*VDU_B+1*VDU_C。由此，本发明实施例可简化了VNFD中的关于容量规格的具体内容，VNFD中的容量规格一栏可不再需要VDU列表的信息。其中，VDU_A、VDU_B和VDU_C代表不同类型的VDU。
步骤S1112，所述网元管理设备根据所述资源配置信息中的VDU与软件的对应关系，确定所述VDU列表中各VDU各自对应的软件。
步骤S1113，所述网元管理设备根据所述资源配置信息中的VDU与虚拟机(Virtual Machine，VM)的对应关系，确定各VDU各自所需的VM规格。具体实现中，不同类型的VDU所需的VM规格可不相同。举例如下：
1个VDU_A所需的VM规格包括：1个VM_A，对应4核vCPU+8G内存+2G带宽；
1个VDU_B所需的VM规格包括：1个VM_B，对应4核vCPU+16G内存+2G带宽；
1个VDU_C所需的VM规格包括：1个VM_C，对应2核vCPU+8G内存+2G带宽。
步骤S1114，所述网元管理设备根据所述资源配置信息中的网络互通关系及各链路的带宽需求，确定各VM各自对链路的需求。
比如，网元管理设备内部存储的资源配置信息中网元的互通关系及带宽需求可如下：
VDU_A与VDU_B互通(也即VM_A与VM_B是互通的)，带宽需要为2G；
VDU_B与VDU_C互通(也即VM_B与VM_C是互通的)，带宽需要为2G；
则步骤S1114确定的链路需求可为VM_A与VM_B之间需建立虚拟链路，VM_B与VM_C之间需建立虚拟链路。
具体实现中，步骤S1112与S1113-S1114之间可无先后执行顺序。
具体实现中，在其他实施例中，步骤S111，还可包括步骤S1115，所述网元管理设备根据所述资源配置信息中的VDU隔离度和亲和度，确定VM的亲和度和隔离度。比如，网元管理设备存储的资源配置信息中的VDU隔离度和亲和度可如下：
VDU_A与VDU_C的VM部署在相同的物理服务器上；
同类型的VDU部署在不同的物理服务器上。
因此，在步骤S11115即可根据网元管理设备存储的VDU隔离度和亲和度确定各VM的亲和度和隔离度为VM_A与VM_C部署在相同的物理服务器上，多个VM_A部署在不同的物理服务器上，比如，VM_1、VM_2的类型均是VM_A，但是在部署时，VM_1和VM_2需部署在不同的物理服务器上。
具体实现中，当接收到实例化VNF请求消息之后，本发明实施例可在网元管理设备中为所述实例化VNF请求消息所请求建立的VNF实例创建实例标识，并建立所述VNF实例的版本号、类型及容量规格与所述实例标识的对应关系。
步骤S112，所述网元管理设备根据所述确定出的资源来实例化VNF。
具体实现中，参考图4，作为一种实施例，步骤S112具体可包括如下步骤：
步骤S1120，所述网元管理设备根据所述确定的VDU列表生成部署VDU的命令列表。
具体实现中，本发明实施例的网元管理设备内存储了各类型的配置命令， 比如，部署命令ADD VDU；删除命令RMV VDU等。因此，在步骤S1120，网元管理设备可为确定的VDU列表中的各VDU分别生成部署命令，整个VDU列表中的各VDU生成的部署命令集合即形成VDU部署命令列表。作为一种示例，本发明实施例可生成如下的VDU部署命令列表(对应的VDU列表为1*VDU_A+1*VDU_B+1*VDU_C)：
ADD VDU：VDU_1(表示部署的VDU的编号)，VDU_A(表示部署的VDU类型)；
ADD VDU：VDU_2(表示部署的VDU的编号)，VDU_B(表示部署的VDU类型)；
ADD VDU：VDU_3(表示部署的VDU的编号)，VDU_C(表示部署的VDU类型)；
步骤S1121，所述网元管理设备在执行所述命令列表中的VDU的部署命令过程中，向VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置。
具体实现中，在步骤S1121可根据步骤S1110-S1115流程所确定的VM资源及链路资源向VIM请求分配对应的资源，以及为所述VDU列表中的各VDU加载软件及进行基础数据配置。具体实现中，所述进行基础数据配置是指将为各VDU加载的软件及各VDU分配的虚拟机VM资源的分配结果及各VDU的类型等信息记录到所述网元管理设备中，并将各VDU与包括这些VDU的VNF实例的标识进行关联，作为后续可用的资源配置信息，以便所述网元管理设备基于这些基础数据配置对VNF实例及实例中的VDU进行管理。由此可见，本发明实施例在执行VNF实例化的过程中即自动完成基础数据配置，解决了现有技术需要手动在网元管理设备中进行基础数据配置的技术问题。
作为一种实施例，所述网元管理设备可逐一执行所述命令列表中的每一个VDU的部署命令，在执行每一个VDU的部署命令过程中，向所述VIM请求分配所述每一个VDU所需的虚拟机VM资源，并根据所述VIM为所述每一个VDU 分配的资源为所述一个VDU加载软件并根据所述VIM对每一个VDU分配所需的虚拟机VM资源的分配结果，为所述每一个VDU加载软件并进行基础数据配置。
当然作为其他的实施方式，在步骤S1121，本发明实施例可一次性向VIM请求所述VDU列表中所有VDU所需的VM资源及链路资源，并在请求完之后，一次性为各VDU加载软件及进行基础数据配置。
步骤S1122，所述网元管理设备在所述VDU列表中所有VDU的部署命令执行完之后，通知所述VNFM所述实例化VNF完成。
具体实现中，作为一种实施例，本发明实施例在步骤S1120之前，还可包括如下步骤：
所述网元管理设备经由VNF管理器(Virtual Network Function Manager，VNFM)向网络功能虚拟化编排器(Network Functions Virtualization Orchestrator，NFVO)发起实例化VNF的操作授权请求，并经所述VNFM从所述NFVO接收所述实例化VNF的操作授权的响应，所述实例化VNF的操作授权的响应是所述NFVO确定存在执行所述实例化VNF所需的资源并请求所述VIM预留相应资源后发送的；
在此情形下，步骤S1121中所述网元管理设备向VIM请求分配所述VDU列表中各VDU所需的VM资源及链路资源，具体包括：
所述网元管理设备经由所述VNFM将所述请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。
具体实现中，作为另一种实施例，步骤S1121中所述网元管理设备向VIM请求分配所述VDU列表中各VDU所需的VM资源，具体包括：
所述网元管理设备依次经由VNFM和NFVO将请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。该实施例与上一实施例的不同在于，由NFVO直接向VIM发起资源分配请求，而上一实施例在发起资源分配请求之前还包括操作授权请求及响应确定网络中有相关资源。
由上可见，本发明实施例中网元管理设备接收VNF管理器VNFM发送的实 例化VNF请求消息；并根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源；以及根据所述确定出的资源来实例化VNF。本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行实例化VNF的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了实例化VNF的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行实例化VNF时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来实例化VNF的具体流程也可不同。
相应的，图5为本发明实施例的VNF实例扩容的方法的流程示意图。本发明实施例中VNF实例扩容可独立于上述实例化VNF的流程执行，或在本发明前述的VNF实例化流程之后进行。如图5所述，其可包括：
步骤S210，网元管理设备获取VNF上报的性能参数。
具体实现中，本发明实施例获取的VNF上报的性能参数可包括CPU占用率，或者其他任意可通过VNF上报的性能参数。
步骤S211，所述网元管理设备根据所述上报的性能参数及所述网元管理设备中存储的资源配置信息，确定执行VNF实例扩容所需部署的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源。
具体实现中，本发明实施例可在网元管理设备内存储的资源配置信息中可维护一份VNF实例扩缩容策略信息，或者所述资源配置信息中可分别维护一份VNF实例扩容策略信息及一份VNF实例缩容策略信息，其中所述VNF实例扩容策略可包括：当上一次上报的性能参数低于第二阈值，本次上报的性能参数大于或等于第二阈值时，执行VNF实例扩容，即在已有的VNF实例容量规格的基础上再部署一个或多个指定类型的VDU；所述VNF实例缩容策略可包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与 一缓冲参数的差值时，执行VNF实例缩容，终止VNF实例中一个或多个指定类型的VDU。
例如：假设VNF上报的性能参数是CPU占用率(当然具体实现中可为其他性能参数)，则本发明实施例的资源配置信息中的VNF实例扩缩容策略信息具体可包括(前2条为VNF实例扩容策略，后2条为VNF实例缩容策略)：
1.当前一次上报的CPU占用率低于20％，本次上报的CPU占用率>＝20％时，扩容一个类型为A的VDU；
2.当前一次上报的CPU占用率低于60％(对应上述的第二阈值)，本次上报的CPU占用率>＝60％时，扩容一个类型为B的VDU；
3.当前一次上报的CPU占用率高于60％，本次上报的CPU占用率<(60％-5％)时，缩容一个类型为A的VDU；
4.当前一次上报的CPU占用率高于20％(对应上述的第一阈值)，本次上报的CPU占用率<(20％-5％(对应上述的缓冲参数))时，缩容一个类型为B的VDU。
这样，在步骤S211，当VNF上报的CPU的占用率符合上述4个策略中的第1个策略或第2个策略时，可确定执行VNF实例扩容所需部署的VDU列表，比如，当满足上述策略1时，在所述VNF原有实例的基础上确定需再部署一个类型为A的VDU。
具体实现中，本发明实施例在网元管理设备中存储的资源配置信息中还可包括容量规格与虚拟部署单元VDU的配比关系、VDU与软件的对应关系、VDU与虚拟机的对应关系、或网络互通关系及各链路的带宽需求等信息。由此，本发明实施例，在步骤S211，当确定需部署的VDU列表之后，可进一步根据网元管理设备中存储的资源配置信息进一步确定所述VDU列表中各VDU所需的虚拟机VM资源。其中，所述虚拟机VM资源可包括：各VDU所需的虚拟机VM规格、以及各VM对链路的需求等。进一步，作为补充实施例，本发明实施例所确定的资源还可包括：各VDU对应的软件及各VM的亲和度和隔离度。
作为一种实施方式，如图6所示，本发明实施例步骤S211，具体可包括：
步骤S2110，所述网元管理设备根据所述上报的性能参数及所述网元管理设备中存储的资源配置信息中的VNF实例扩容策略信息，确定执行VNF实例扩容所需部署的VDU列表。具体如何确定VDU列表参考前面的描述，在此不进行赘述。
步骤S2111，所述网元管理设备根据所述资源配置信息中的VDU与软件的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自对应的软件。
步骤S2112，所述网元管理设备根据所述资源配置信息中的VDU与虚拟机(Virtual Machine，VM)的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自所需的VM规格。具体实现中，不同类型的VDU所需的VM规格可不相同。举例如下：
1个VDU_A所需的VM规格包括：1个VM_A，对应4核vCPU+8G内存+2G带宽；
1个VDU_B所需的VM规格包括：1个VM_B，对应4核vCPU+16G内存+2G带宽；
1个VDU_C所需的VM规格包括：1个VM_C，对应2核vCPU+8G内存+2G带宽。
步骤S2113，所述网元管理设备根据所述资源配置信息中的网络互通关系及各链路的带宽需求，所述VM各自对链路的需求。
比如，网元管理设备内部存储的资源配置信息中网元的互通关系及带宽需求可如下：
VDU_A与VDU_B互通(也即VM_A与VM_B是互通的)，带宽需要为2G；
VDU_B与VDU_C互通(也即VM_B与VM_C是互通的)，带宽需要为2G；
具体实现中，步骤S2111与S2112-S2113之间可无先后执行顺序。具体实现中，在其他实施例中，步骤S211，还可包括步骤S2114，所述网元管理设备根据所述资源配置信息中的VDU隔离度和亲和度，确定VM的亲和度和隔离度。比如，网元管理设备存储的资源配置信息中的VDU隔离度和亲和度可如下：
VDU_A与VDU_C的VM部署在相同的物理服务器上；
同类型的VDU部署在不同的物理服务器上。
因此，在步骤S2114即可根据网元管理设备存储的VDU隔离度和亲和度确定各VM的亲和度和隔离度为VM_A与VM_C部署在相同的物理服务器上，多个VM_A部署在不同的物理服务器上，比如，VM_1、VM_2的类型均是VM_A，但是在部署时，VM_1和VM_2需部署在不同的物理服务器上。
作为一种可替代方案，具体实现中，如图7所示，本发明实施例的步骤S210-S211可替换为如下流程：
步骤S220，网元管理设备从VMFM接收VNF实例扩容请求；
步骤S221，所述网元管理设备根据所述接收的VNF实例扩容请求和所述网元管理设备存储的资源配置信息，确定执行VNF实例扩容所需部署的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源。
其中，如图8所示，作为一种实施例，步骤S221，可包括：
步骤S2210，所述网元管理设备根据所述VNF实例扩容请求中携带的VNF实例标识及扩容后的容量规格，确定要部署的VDU列表；
步骤S2211，所述网元管理设备根据所述资源配置信息中的VDU与软件的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自对应的软件；
步骤S2212，所述网元管理设备根据所述资源配置信息中的VDU与虚拟机(Virtual Machine，VM)的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自所需的VM规格；
步骤S2213，所述网元管理设备根据所述资源配置信息中的网络互通关系及各链路的带宽需求，所述VM各自对链路的需求。
具体实现中，步骤S2211与S2212-S2213之间可无先后执行顺序。
除上述步骤之外，还可包括步骤S2114，所述网元管理设备根据所述资源配置信息中的VDU隔离度和亲和度，确定VM的亲和度和隔离度。
其中，步骤S2211-步骤S2214分别与步骤S2111-步骤S2114相同，因此， 在此，对步骤S2211-步骤S2214具体实现不再赘述。
步骤S212，所述网元管理设备根据所述确定的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源，执行VNF实例扩容。
具体实现中，参考图9，作为一种实施例，步骤S212具体可包括如下步骤：
步骤S2120，所述网元管理设备根据所述确定的VDU列表生成部署VDU的命令列表。假设步骤S211或S221确定的VDU列表中包括1个类型为A的VDU(简化为：1*VDU_A)。
具体实现中，本发明实施例网元管理设备内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，在步骤S2120，网元管理设备可为确定的VDU列表中的各VDU分别生成部署命令，整个VDU列表中的各VDU生成的部署命令集合即形成VDU部署命令列表。作为一种示例，当确定的VDU列表为：1*VDU_A时，本发明实施例可生成如下的VDU部署命令列表：
ADD VDU：VDU_1(表示部署的VDU的编号)，VDU_A(表示部署的VDU类型)；
步骤S2121，所述网元管理设备在执行所述命令列表中的VDU的部署命令过程中，向VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置。
具体实现中，所述进行基础数据配置是指将为各VDU加载的软件及各VDU分配的虚拟机VM资源的分配结果及各VDU的类型等信息记录到所述网元管理设备中，并将各VDU与包括这些VDU的VNF实例的标识进行关联，作为后续可用的资源配置信息，以便所述网元管理设备基于这些基础数据配置对VNF实例及实例中的VDU进行管理。由此可见，本发明实施例在执行VNF实例扩容的过程中即自动完成基础数据配置，解决了现有技术需要手动在网元管理设备中进行基础数据配置的技术问题。
作为一种实施例，在步骤S2121，所述网元管理设备可逐一执行所述命令列 表中的每一个VDU的部署命令，在执行每一个VDU的部署命令过程中，向所述VIM请求分配所述每一个VDU所需的虚拟机VM资源，并根据所述VIM为所述每一个VDU分配的资源为所述一个VDU加载软件并根据所述VIM对每一个VDU分配所需的虚拟机VM资源的分配结果，为所述每一个VDU加载软件并进行基础数据配置。
当然作为其他的实施方式，在步骤S2121，本发明实施例可一次性向VIM请求所述VDU列表中所有VDU所需的VM资源及链路资源，并在请求完之后，一次性为各VDU加载软件及进行基础数据配置。
步骤S2122，所述网元管理设备在所述VDU列表中所有VDU的部署命令执行完之后，通知所述VNFM所述VNF实例扩容完成。
具体实现中，作为一种实施例，本发明实施例在步骤S2120之前，还可包括如下步骤：
所述网元管理设备经由VNF管理器(Virtual Network Function Manager，VNFM)向网络功能虚拟化编排器(Network Functions Virtualization Orchestrator，NFVO)发起实例化VNF的操作授权请求，并经所述VNFM从所述NFVO接收所述实例化VNF的操作授权的响应，所述实例化VNF的操作授权的响应是所述NFVO确定存在执行所述实例化VNF所需的资源并请求所述VIM预留相应资源后发送的；
在此情形下，步骤S2121中所述网元管理设备向VIM请求分配所述VDU列表中各VDU所需的VM资源，具体包括：
所述网元管理设备经由所述VNFM将所述请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。
具体实现中，作为另一种实施例，步骤S2121中所述网元管理设备向VIM请求分配所述VDU列表中各VDU所需的VM资源，具体包括：
所述网元管理设备依次经由VNFM和NFVO将请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。该实施例与上一实施例的不同在于，由NFVO直接向VIM发起资源分配请求，而上一实施例在发起资源分 配请求之前还包括操作授权请求及响应确定网络中有相关资源。
由上可见，本发明实施例中网元管理设备获取VNF上报的性能参数或接收的VNF实例扩容请求；根据所述上报的性能参数或接收的VNF实例扩容请求及所述网元管理设备中存储的资源配置信息，确定执行VNF实例扩容所需部署的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源；以及根据所述确定出的资源来执行VNF实例扩容。本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行VNF实例扩容的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了VNF实例扩容的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行VNF实例扩容时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来执行VNF实例扩容的具体流程也可不同。
图10为本发明实施例的VNF实例缩容的方法的流程示意图。本发明实施例中VNF实例缩容可独立于上述实例化VNF的流程执行，或在本发明前述的VNF实例化流程之后进行。如图10所述，其可包括：
步骤S310，所述网元管理设备获取VNF上报的性能参数；
具体实现中，本发明实施例获取的VNF上报的性能参数可包括CPU占用率，或者其他任意可通过VNF上报的性能参数。
步骤S311，所述网元管理设备根据所述上报的性能参数及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，所述要终止的VDU列表中包括一个或多个需终止的VDU，所述资源配置信息中的VNF实例缩容策略包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止一个或多个指定类型的VDU。
具体实现中，本发明实施例可在网元管理设备内存储的资源配置信息中可 维护一份VNF实例扩缩容策略信息，或者所述资源配置信息中可分别维护一份VNF实例扩容策略信息及一份VNF实例缩容策略信息，其中所述VNF实例扩容策略可包括：当上一次上报的性能参数低于第二阈值，本次上报的性能参数大于或等于第二阈值时，执行VNF实例扩容，即在已有的VNF实例容量规格的基础上再部署一个或多个指定类型的VDU；所述VNF实例缩容策略可包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止VNF实例中一个或多个指定类型的VDU。
例如：
假设VNF上报的性能参数是CPU占用率(当然具体实现中可为其他性能参数)，则本发明实施例的资源配置信息中的VNF实例扩缩容策略信息具体可包括(前2条为VNF实例扩容策略，后2条为VNF实例缩容策略)：
1.当前一次上报的CPU占用率低于20％，本次上报的CPU占用率>＝20％时，扩容一个类型为A的VDU；
2.当前一次上报的CPU占用率低于60％(对应上述的第二阈值)，本次上报的CPU占用率>＝60％时，扩容一个类型为B的VDU；
3.当前一次上报的CPU占用率高于60％，本次上报的CPU占用率<(60％-5％)时，缩容一个类型为A的VDU；
4.当前一次上报的CPU占用率高于20％(对应上述的第一阈值)，本次上报的CPU占用率<(20％-5％(对应上述的缓冲参数))时，缩容一个类型为B的VDU。
这样，在步骤S311，当VNF上报的CPU的占用率符合上述4个策略中的第3个策略或第4个策略时，可确定执行VNF实例缩容所需终止的VDU列表，比如，当满足上述策略3时，确定在所述VNF原有实例的基础上需终止一个类型为A的VDU。
具体实现中，本发明实施例在网元管理设备中存储的资源配置信息中还可包括容量规格与虚拟部署单元VDU的配比关系、VDU与软件的对应关系、VDU 与虚拟机的对应关系、网络互通关系及各链路的带宽需求等信息。由此，本发明实施例，在步骤S311，当确定需终止的VDU列表之后，可进一步根据网元管理设备中存储的资源配置信息进一步确定所述VDU列表中各VDU所需释放的虚拟机VM资源及各VDU需删除的软件。其中，所述虚拟机VM资源可包括：各VDU所需的虚拟机VM规格、以及各VM对链路的需求等。进一步，作为补充实施例，本发明实施例在步骤S311还可确定需终止的与所述确定的需终止的VDU具有亲和度和隔离度关系的VDU。
作为一种实施方式，如图11所示，本发明实施例步骤S311，具体可包括：
步骤S3110，所述网元管理设备根据所述上报的性能参数及所述网元管理设备中存储的资源配置信息中的VNF实例缩容策略信息，确定执行VNF实例缩容所需终止的VDU列表。具体如何确定参考前面的描述，在此不进行赘述。
步骤S3111，所述网元管理设备根据所述资源配置信息中各VDU的基础数据配置信息，确定所述需终止的VDU列表中各VDU需删除的软件、需释放的VM规格、以及需释放的与VM相关的链路。
具体实现中，不同类型的VDU所对应的VM规格可不相同。举例如下：
1个VDU_A所需的VM规格包括：1个VM_A，对应4核vCPU+8G内存+2G带宽；
1个VDU_B所需的VM规格包括：1个VM_B，对应4核vCPU+16G内存+2G带宽；
1个VDU_C所需的VM规格包括：1个VM_C，对应2核vCPU+8G内存+2G带宽。
比如，当确定需终止1个VDU_A时，可确定需删除1个VM_A，删除该VM_A具体表现为需释放4核vCPU+8G内存+2G带宽。
再如，假设网元管理设备内部存储的资源配置信息中网元的互通关系及带宽需求如下：
VDU_A与VDU_B互通(也即VM_A与VM_B是互通的)，带宽需要为2G；
VDU_B与VDU_C互通(也即VM_B与VM_C是互通的)，带宽需要为2G；
因此，本发明实施例在步骤S3111即可根据上述互通关系，确定需释放VM_A与VM_B之间的链路，以及释放VM_B与VM_C之间的链路。
具体实现中，在删除各VDU对应的软件、需释放的VM规格、以及需释放的与VM相关的链路之后，还可删除所述网元管理单元中配置的VDU的基础数据，比如，解除所述VDU与VNF实例的标识的关联关系，删除存储的所述VDU与软件的对应关系信息，删除所述VDU与VM规格的对应关系，以及删除所述VDU的VM的相关的链路信息等。
具体实现中，在其他实施例中，步骤S311，还可包括步骤S3112，所述网元管理设备根据所述资源配置信息中的VM隔离度和亲和度，确定需终止的与VDU列表中VDU具有亲和度和隔离度关系的VDU。比如，网元管理设备存储的资源配置信息中的VM隔离度和亲和度可如下：
VM_A与VM_C部署在相同的物理服务器上；
同类型的VM部署在不同的物理服务器上。
因此，比如假设确定需终止VDU_A，在终止位于物理服务器1上的类型为VDU_A的编号为VDU_1的VDU时，还要终止位于物理服务器1上的类型为VDU_C的编号为VDU_3的VDU。
作为一种可替代的方案，如图12，具体实现中，步骤S310-S311可由如下流程替换：
步骤S320，网元管理设备从VNFM接收缩容VNF实例请求；
步骤S321，所述网元管理设备根据所述接收的缩容VNF实例请求和所述网元管理设备存储的资源配置信息，确定执行VNF实例缩容所需终止的VDU列表及所述VDU列表中各VDU所对应的需释放的VM资源。
其中，如图13所示，作为一种实施例，步骤S321，可包括：
步骤S3210，所述网元管理设备根据所述接收的缩容VNF实例请求中携带的VNF实例标识及缩容后的容量规格，确定执行VNF实例缩容需终止的VDU列表；
步骤S3211，所述网元管理设备根据所述资源配置信息中各VDU的基础数 据配置信息，确定所述需终止的VDU列表中各VDU需删除的软件、需释放的VM规格、以及需释放的与VM相关的链路。
具体实现中，在其他实施例中，步骤S311，还可包括步骤S3212，所述网元管理设备根据所述资源配置信息中的VM隔离度和亲和度，确定需终止的与VDU列表中VDU具有亲和度和隔离度关系的VDU。
其中，步骤S3211-S3212与步骤S3111-S3112分别相同，在此，对步骤S3211-S3212的具体实现不进行赘述。
步骤S312，所述网元管理设备根据所述确定的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例缩容。
作为一种实施例，如图14所示，步骤S312包括：
步骤S3121，所述网元管理设备根据确定的要终止的VDU列表生成删除VDU的命令列表。假设步骤S311确定的列表中包括1个类型为A的VDU(简化为：1*VDU_A)。
具体实现中，本发明实施例在网元管理设备内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，在步骤S3120，网元管理设备可为确定的VDU列表中的各VDU分别生成删除命令，整个VDU列表中的各VDU生成的删除命令集合即形成VDU删除命令列表。作为一种示例，当确定的VDU列表为：1*VDU_A时，本发明实施例可生成如下的VDU删除命令列表：
RMV VDU：VDU_1(表示终止的VDU的编号)，VDU_A(表示终止的VDU类型)；
步骤S3122，所述网元管理设备在执行所述删除VDU的命令列表中的VDU删除命令的过程中，卸载各VDU的软件，并通过VNFM向所述VIM请求释放各VDU相关的虚拟机VM资源，并将所述网元管理设备中与所述VDU相关的基础数据删除。
步骤S3123，所述网元管理设备在执行完所述删除VDU的命令列表中的所有VDU删除命令后，通知VNFM所述VNF实例化缩容完成。
由上可见，本发明实施例中网元管理设备获取VNF上报的性能参数；根据所述上报的性能参数及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM规格资源，所述要终止的VDU列表中包括一个或多个需终止的VDU，所述资源配置信息中的VNF实例缩容策略包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止一个或多个指定类型的VDU；以及根据所述确定的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例缩容。本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行VNF实例缩容的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了VNF实例缩容的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行VNF实例缩容时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来执行VNF实例缩容的具体流程也可不同。另外，本发明实施例的VNF实例缩容策略中通过一缓冲参数来降低了VNF的乒乓弹性伸缩。
图15为本发明实施例的终止VNF实例的方法的流程示意图。本发明实施例中终止VNF实例的流程可独立于上述实例化VNF的流程执行，或在本发明前述的VNF实例化流程之后进行。如图15所述，其可包括：
步骤S410，网元管理设备接收VNF管理器VMFM发送的终止VNF实例请求消息。
步骤S411，所述网元管理设备根据所述终止VNF实例请求消息和所述网元管理设备中存储的资源配置信息确定执行终止VNF实例需要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源。
作为一种实施方式，如图16所示，步骤S411，具体可包括：
步骤S4110，所述网元管理设备根据所述终止VNF实例请求消息中携带的VNF实例标识和所述网元管理设备中存储的资源配置信息，确定要终止的VNF实例及需终止的VDU列表；
步骤S4111，所述网元管理设备根据所述资源配置信息中各VDU的基础数据配置信息，确定所述需终止的VDU列表中各VDU需删除的软件、需释放的VM规格、以及需释放的与VM相关的链路。
具体实现中，不同类型的VDU所对应的VM规格可不相同。举例如下：
1个VDU_A所需的VM规格包括：1个VM_A，对应4核vCPU+8G内存+2G带宽；
1个VDU_B所需的VM规格包括：1个VM_B，对应4核vCPU+16G内存+2G带宽；
1个VDU_C所需的VM规格包括：1个VM_C，对应2核vCPU+8G内存+2G带宽。
因此，本发明实施例在步骤S4111即可根据VDU与VM规格的对应关系，确定需删除的VM。比如，当确定需终止VDU列表为1*VDU_A+1*VDU_B+1*VDU_C，可确定需删除1个VM_A，删除该VM_A具体表现为需释放4核vCPU+8G内存+2G带宽；需删除1个VM_B，删除该VM_B具体表现为需释放4核vCPU+16G内存+2G带宽；需删除1个VM_C，删除该VM_C具体表现为需释放2核vCPU+8G内存+2G带宽。
再如，网元管理设备内部存储的资源配置信息中网元的互通关系及带宽需求可如下：
VDU_A与VDU_B互通(也即VM_A与VM_B是互通的)，带宽需要为2G；
VDU_B与VDU_C互通(也即VM_B与VM_C是互通的)，带宽需要为2G；
因此，本发明实施例在步骤S4111即可根据上述互通关系，确定需释放VM_A与VM_B之间的链路，以及释放VM_B与VM_C之间的链路。具体实现中，在其他实施例中，步骤S411，还可包括步骤S4112，所述网元管理设备根据所述资源配置信息中的VM隔离度和亲和度，确定需终止的与VDU列表中 VDU具有亲和度和隔离度关系的VDU。比如，网元管理设备存储的资源配置信息中的VM隔离度和亲和度可如下：
VM_A与VM_C部署在相同的物理服务器上；
同类型的VM部署在不同的物理服务器上。
因此，比如假设确定需终止VDU_A，在终止位于物理服务器1上的类型为VDU_A的编号为VDU_1的VDU时，还要终止位于物理服务器1上的类型为VDU_C的编号为VDU_3的VDU。
步骤S412，所述网元管理设备根据所述确定出的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例终止。
作为一种实施例，如图17所示，步骤S412包括：
步骤S4121，所述网元管理设备根据确定的要终止的VDU列表生成删除VDU的命令列表。假设步骤S411确定的列表为1*VDU_A+1*VDU_B+1*VDU_C。
具体实现中，本发明实施例网元管理设备内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，在步骤S4120，网元管理设备可为确定的VDU列表中的各VDU分别生成删除命令，整个VDU列表中的各VDU生成的删除命令集合即形成VDU删除命令列表。
作为一种示例，当步骤S411确定的列表为1*VDU_A+1*VDU_B+1*VDU_C时，本发明实施例可生成如下的VDU删除命令列表：
RMV VDU：VDU_1(表示终止的VDU的编号)，VDU_A(表示终止的VDU类型)；
RMV VDU：VDU_2(表示终止的VDU的编号)，VDU_B(表示终止的VDU类型)；
RMV VDU：VDU_3(表示终止的VDU的编号)，VDU_C(表示终止的VDU类型)。
步骤S4122，所述网元管理设备在执行所述删除VDU的命令列表中的VDU删除命令的过程中，卸载各VDU的软件，并通过VNFM向所述VIM请求释放 各VDU相关的虚拟机VM资源，并将所述网元管理设备中与所述VDU相关的基础数据删除。比如，解除所述VDU与VNF实例的标识的关联关系，删除存储的所述VDU与软件的对应关系信息，删除所述VDU与VM规格的对应关系，以及删除所述VDU的VM的相关的链路信息等。步骤S4123，所述网元管理设备在执行完所述删除VDU的命令列表中的所有VDU删除命令后，通知VNFM所述VNF实例化终止完成。
由上可见，本发明实施例中网元管理设备接收VNF管理器VMFM发送的终止VNF实例请求消息；根据所述终止VNF实例请求消息和所述网元管理设备中存储的资源配置信息确定执行终止VNF实例需要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，所述要终止的VDU列表中包括一个或多个需终止的VDU，以及根据所述确定的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例终止。本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行VNF实例终止的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了VNF实例终止的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行VNF实例终止时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来执行VNF实例终止的具体流程也可不同。
由此可见，本发明实施例其相对于现有技术中VNFM(可为图1所示架构中的VNFM)完全依赖VNFD执行相应的VNF生命周期管理具有如下优势：
不再要求各Vendor提供的VNFD覆盖各个生命周期管理全部过程，也不需要VNFD做到标准化，不同的Vendor的网元管理设备之间能体现各Vendor的差异性。并且，当网元管理设备收到一个生命周期(Lifecycle)管理的操作后，不再需要严格根据VNFD中相应参数或流程、脚本的指示，一步一步完成相应 的动作，使得VNF的生命周期管理过程非常灵活性。
经过本发明实施例的方法之后，最终VNFD可简化如下：
简化方式1：VNFD中仅保留vendor(对应表1中参数列的vendor及其对应的参数值)、VNF类型(对应表1中参数列的VNF id及其对应的参数值)、版本(对应表1中参数列的VNF版本及其对应的参数值)、容量规格(对应表1中参数列的容量及其对应的参数值)等基本信息。而现有技术VNFD中的其他信息均不需要保留。
简化方式2：将整个VNFD去除，但需将NSD模板做适当修改，确保原来通过关联VNFD能得到的Vendor、VNF类型、版本信息，容量规格等能够在NSD中体现。
总之，不论采用哪种简化方式，只要执行本发明实施例的方法时NFVO->VNFM的接口包含如下信息即可：
Vendor+VNF类型：用于VNFM判断将请求转发到正确的网元管理设备；
VNF类型、VNF版本号、容量规格：由VNFM转发到网元管理设备，以使网元管理设备用此决策实例化哪个版本的VNF，按多大的规格来部署。
具体实现中，当执行VNF生命周期管理的实体为LMU时，本发明实施例还需VNFD或NSD中需保留所述LMU的参数和脚本，进而使得本发明实施例在通过LMU执行VNF生命周期管理之前，能首先将LMU部署起来。具体实现中，LMU的部署流程与现有技术中通过VNFD部署VDU流程一致，在此不进行赘述。
相应的，本发明实施例还提供了可用于实施上述各方法的网元管理设备。下面结合具体实施例和附图对本发明实施例的网元管理设备的结构组成进行举例说明。
图18为本发明实施例的网元管理设备的一实施例的结构组成示意图。如图18所示，其可包括：接收模块10、确定模块11以及实例化模块12，其中：
接收模块10，用于接收VNF管理器VNFM发送的实例化VNF请求消息。
具体实现中，当接收到实例化VNF请求消息之后，本发明实施例可在网元 管理设备中为所述实例化VNF请求消息所请求建立的VNF实例创建实例标识，并建立所述VNF实例的版本号、类型及容量规格与所述实例标识的对应关系。具体实现中，本发明实施例的所述网元管理设备与所述VNF由同一供应商提供，可包括但不限于网元管理系统(Element Management System，EMS)或VNF本地管理单元(Local Management Unit，LMU)。所述网元管理设备的接收模块10可通过与VNFM之间的Ve-Vnfm接口从所述VNFM接收实例化VNF请求消息。
确定模块11，用于根据所述接收模块接收的所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源。
具体实现中，所述确定模块11具体可根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定实例化VNF所需的虚拟部署单元VDU列表，以及所述VDU列表中各VDU所需虚拟机VM资源。其中，所述虚拟机VM资源可包括：各VDU所需的虚拟机VM规格、以及各VM对链路的需求等。进一步，作为补充实施例，本发明实施例确定模块11所确定的执行实例化VNF所需的资源还可包括：各VDU对应的软件及各VM的亲和度和隔离度。
具体实现中，参考图19，作为一种实施例，确定模块11可包括：规格确定模块110、VDU列表确定模块111、软件确定模块112、VM规格确定模块113以及链路确定模块114，其中：
规格确定模块110，用于根据所述接收模块10接收的实例化VNF的请求确定要部署的VNF的类型、版本号及容量规格。
具体实现中，规格确定模块110可根据所述实例化VNF请求中携带的VNF的信息模型描述模版VNFD索引信息和容量规格，到对应的VNFD中查询要部署的VNF的类型，版本号及容量规格。如何通过VNFD索引找到对应的VNFD及查询到要部署的VNF的类型，版本号，容量规格的示例详见方法步骤S1110，在此不进行赘述。
或者，规格确定模块110可根据所述实例化VNF请求携带的VNF的类型、版本号及容量规格确定要部署的VNF的类型、版本号及容量规格。如何根据请 求携带的VNF的类型、版本号及容量规格确定要部署的VNF的类型、版本号及容量规格的示例，详见方法流程S1110，在此不进行赘述。
具体实现中，所述实例化VNF请求携带的VNF的类型、版本号及容量规格注册在所述VNFD中或注册在网络业务的信息模型描述模版NSD中。
VDU列表确定模块111，用于根据所述资源配置信息中的容量规格与虚拟部署单元VDU的配比关系，确定所述要部署的VNF所需的VDU列表。
本发明实施例中，VDU列表确定模块111确定所要部署的VNF所需的VDU列表的具体手段可与步骤S1111相同，在此不进行赘述。
软件确定模块112，用于根据所述资源配置信息中的VDU与软件的对应关系，确定所述VDU列表中各VDU各自对应的软件。
VM规格确定模块113，用于根据所述资源配置信息中的VDU与虚拟机(Virtual Machine，VM)的对应关系，确定各VDU各自所需的VM规格。具体实现中，不同类型的VDU所需的VM规格可不相同。关于VDU所需的VM规格的示例可为步骤S1113所举示例，在此不进行赘述。
链路确定模块114，用于根据所述资源配置信息中的网络互通关系及各链路的带宽需求，确定各VM各自对链路的需求。
具体实现中，链路确定模块114确定各VM各自对链路的需求的手段可与方法步骤S1114相同，在此不进行赘述。
具体实现中，在其他实施例中，确定模块11还可包括隔离亲和度确定模块115，用于根据所述资源配置信息中的VDU隔离度和亲和度，确定VM的亲和度和隔离度。
隔离亲和度确定模块115确定各VM各自对链路的需求的手段可与方法步骤S1115相同，在此不进行赘述。
实例化模块12，用于根据所述确定模块所确定出的资源来实例化VNF。
具体实现中，参考图20，作为一种实施例，实例化模块12具体可包括：生成模块120、部署模块121以及通知模块122，其中：
生成模块120，用于根据所述确定的VDU列表生成部署VDU的命令列表。
具体实现中，本发明实施例的网元管理设备内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，生成模块120可为确定的VDU列表中的各VDU分别生成部署命令，整个VDU列表中的各VDU生成的部署命令集合即形成VDU部署命令列表。本实施例可生成的VDU部署命令列表示例可为步骤S1120所举示例，在此不进行赘述。
部署模块121，用于执行所述生成模块120生成的部署VDU的命令列表中的VDU部署命令，并在执行所述命令列表中的VDU的部署命令过程中，向VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置。
具体实现中，部署模块121可根据确定模块11所确定的VM资源及链路资源向VIM请求分配对应的资源，以及为所述VDU列表中的各VDU加载软件及进行基础数据配置。具体实现中，所述进行基础数据配置是指将为各VDU加载的软件及各VDU分配的虚拟机VM资源的分配结果及各VDU的类型等信息记录到所述网元管理设备中，并将各VDU与包括这些VDU的VNF实例的标识进行关联，作为后续可用的资源配置信息，以便所述网元管理设备基于这些基础数据配置对VNF实例及实例中的VDU进行管理。由此可见，本发明实施例在执行VNF实例化的过程中即自动完成基础数据配置，解决了现有技术需要手动在网元管理设备中进行基础数据配置的技术问题。作为一种实施例，部署模块121可逐一执行所述命令列表中的每一个VDU的部署命令，在执行每一个VDU的部署命令过程中，向所述VIM请求分配所述每一个VDU所需的虚拟机VM资源，并根据所述VIM为所述每一个VDU分配的资源为所述一个VDU加载软件并根据所述VIM对每一个VDU分配所需的虚拟机VM资源的分配结果，为所述每一个VDU加载软件并进行基础数据配置。
当然作为其他的实施方式，部署模块121也可一次性向VIM请求所述VDU列表中所有VDU所需的VM资源及链路资源，并在请求完之后，一次性为各VDU加载软件及进行基础数据配置。
通知模块122，用于在所述部署模块121将所述VDU列表中所有VDU的部署命令执行完之后，通知所述VNFM所述实例化VNF完成。
具体实现中，作为一种实施例，本发明实施例的网元管理设备，还可包括操作授权处理模块，用于经由VNF管理器(Virtual Network Function Manager，VNFM)向网络功能虚拟化编排器(Network Functions Virtualization Orchestrator，NFVO)发起实例化VNF的操作授权请求，并经所述VNFM从所述NFVO接收所述实例化VNF的操作授权的响应，所述实例化VNF的操作授权的响应是所述NFVO确定存在执行所述实例化VNF所需的资源并请求所述VIM预留相应资源后发送的；
在此情形下，部署模块121向VIM请求分配所述VDU列表中各VDU所需的VM资源及链路资源，具体用于经由所述VNFM将所述请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。
具体实现中，作为另一种实施例，部署模块121向VIM请求分配所述VDU列表中各VDU所需的VM资源，具体用于依次经由VNFM和NFVO将请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。该实施例与上一实施例的不同在于，由NFVO直接向VIM发起资源分配请求，而上一实施例在发起资源分配请求之前还包括操作授权请求及响应确定网络中有相关资源。由上可见，本发明实施例的网元管理设备将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行实例化VNF的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了实例化VNF的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行实例化VNF时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来实例化VNF的具体流程也可不同。
作为补充，本实施例的网元管理设备还可包括：
获取模块(未图示)，用于获取VNF上报的性能参数；
所述确定模块11还用于根据所述上报的性能参数及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，所述要终止的VDU列表中包括一个或多个需终止的VDU，所述资源配置信息中的VNF实例缩容策略包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止一个或多个指定类型的VDU。
本实施例的网元管理设备还可包括：缩容模块(未图示)，用于根据所述确定模块11确定的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例缩容。
图21为本发明实施例的网元管理设备的另一实施例的结构组成示意图。如图21所示，其可包括：获取模块20、确定模块21以及扩容模块22，其中：
获取模块20，用于获取VNF上报的性能参数。
具体实现中，本发明实施例获取模块20获取的VNF上报的性能参数可包括CPU占用率，或者其他任意可通过VNF上报的性能参数。
确定模块21，用于根据所述VNF上报的性能参数及所述网元管理设备中存储的资源配置信息，确定执行VNF实例扩容所需部署的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源。
具体实现中，本发明实施例可在网元管理设备内存储的资源配置信息中可维护一份VNF实例扩缩容策略信息，或者所述资源配置信息中可分别维护一份VNF实例扩容策略信息及一份VNF实例缩容策略信息，其中所述VNF实例扩容策略可包括：当上一次上报的性能参数低于第二阈值，本次上报的性能参数大于或等于第二阈值时，执行VNF实例扩容，即在已有的VNF实例容量规格的基础上再部署一个或多个指定类型的VDU；所述VNF实例缩容策略可包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止VNF实例中一个或多个指定类型的VDU。
具体实现中，本发明实施例存储的VNF扩缩容策略及VNF上报的性能参数的举例可参考步骤S211的相关描述，在此不进行赘述。
这样，当VNF上报的CPU的占用率符合上述4个策略中的第1个策略或第2个策略时，确定模块21可确定执行VNF实例扩容所需部署的VDU列表，比如，当满足上述策略1时，在所述VNF原有实例的基础上确定需再部署一个类型为A的VDU。
具体实现中，本发明实施例在网元管理设备中存储的资源配置信息中还可包括容量规格与虚拟部署单元VDU的配比关系、VDU与软件的对应关系、VDU与虚拟机的对应关系、或网络互通关系及各链路的带宽需求等信息。由此，本发明实施例，当确定需部署的VDU列表之后，可进一步根据网元管理设备中存储的资源配置信息进一步确定所述VDU列表中各VDU所需的虚拟机VM资源。其中，所述虚拟机VM资源可包括：各VDU所需的虚拟机VM规格、以及各VM对链路的需求等。进一步，作为补充实施例，本发明实施例所确定的资源还可包括：各VDU对应的软件及各VM的亲和度和隔离度。
作为一种实施方式，如图22所示，本发明实施例的确定模块21具体可包括：VDU列表确定模块211、软件确定模块212、VM规格确定模块213以及链路确定模块214，其中：
VDU列表确定模块211，用于根据所述VNF上报的性能参数及所述网元管理设备中存储的资源配置信息中的VNF实例扩容策略信息，确定执行VNF实例扩容所需部署的VDU列表。具体如何确定VDU列表参考前面的描述，在此不进行赘述。
软件确定模块212，用于根据所述网元管理设备中存储的资源配置信息中的VDU与软件的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自对应的软件。
VM规格确定模块213，用于根据所述网元管理设备中存储的资源配置信息中的VDU与虚拟机(Virtual Machine，VM)的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自所需的VM规格。具体实现中，不同 类型的VDU所需的VM规格可不相同。VDU与VM规格的对应关系举例可见步骤S2112的相关描述，在此不进行赘述。
链路确定模块214，用于根据所述网元管理设备中存储的资源配置信息中的网络互通关系及各链路的带宽需求，所述VM各自对链路的需求。
具体实现中，网元管理设备内部存储的资源配置信息中网元的互通关系及带宽需求的示例可如步骤S2113所述，在此不进行赘述。
具体实现中，在其他实施例中，确定模块21，还可包括隔离亲和度确定模块215，用于根据所述网元管理设备中存储的资源配置信息中的VDU隔离度和亲和度，确定VM的亲和度和隔离度。具体实现中，网元管理设备存储的资源配置信息中的VDU隔离度和亲和度关系的示例可如步骤S2114所述，在此不进行赘述。
因此，隔离亲和度确定模块215即可根据网元管理设备存储的VDU隔离度和亲和度确定各VM的亲和度和隔离度为VM_A与VM_C部署在相同的物理服务器上，多个VM_A部署在不同的物理服务器上，比如，VM_1、VM_2的类型均是VM_A，但是在部署时，VM_1和VM_2需部署在不同的物理服务器上。
具体实现中，参考图23，作为一种实施例，扩容模块22具体可包括：
生成模块221，用于根据所述确定模块21确定的VDU列表生成部署VDU的命令列表。假设步骤S211或S221确定的VDU列表中包括1个类型为A的VDU(简化为：1*VDU_A)。
具体实现中，本发明实施例网元管理设备内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，网元管理设备的生成模块221可为确定的VDU列表中的各VDU分别生成部署命令，整个VDU列表中的各VDU生成的部署命令集合即形成VDU部署命令列表。具体实现中，本实施例生成的VDU部署命令列表示例可与步骤S2120所举示例相同，在此不进行赘述。
部署模块222，用于执行所述命令列表中的VDU的部署命令，并在执行所述命令列表中的VDU的部署命令过程中，向VIM请求所述命令列表中各VDU 所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置。
具体实现中，所述进行基础数据配置是指将为各VDU加载的软件及各VDU分配的虚拟机VM资源的分配结果及各VDU的类型等信息记录到所述网元管理设备中作为后续可用的资源配置信息，并将各VDU与包括这些VDU的VNF实例的标识进行关联，作为后续可用的资源配置信息，以便所述网元管理设备基于这些基础数据配置对VNF实例及实例中的VDU进行管理。由此可见，本发明实施例在执行VNF实例扩容的过程中即自动完成基础数据配置，解决了现有技术需要手动在网元管理设备中进行基础数据配置的技术问题。
作为一种实施例，部署模块222可逐一执行所述命令列表中的每一个VDU的部署命令，在执行每一个VDU的部署命令过程中，向所述VIM请求分配所述每一个VDU所需的虚拟机VM资源，并根据所述VIM为所述每一个VDU分配的资源为所述一个VDU加载软件并根据所述VIM对每一个VDU分配所需的虚拟机VM资源的分配结果，为所述每一个VDU加载软件并进行基础数据配置。
当然作为其他的实施方式，部署模块222也可一次性向VIM请求所述VDU列表中所有VDU所需的VM资源及链路资源，并在请求完之后，一次性为各VDU加载软件及进行基础数据配置。
通知模块223，用于在所述部署模块222将所述VDU列表中所有VDU的部署命令执行完之后，通知所述VNFM所述VNF实例扩容完成。
具体实现中，作为一种实施例，本发明实施例，还可包括：
操作授权处理模块(未图示)，用于经由VNF管理器VNFM向网络功能虚拟化编排器NFVO发起VNF实例扩容的操作授权请求，并经所述VNFM从所述NFVO接收所述VNF实例扩容的操作授权的响应，所述VNF实例扩容的操作授权的响应是所述NFVO确定存在执行所述VNF实例扩容所需的资源并请求所述VIM预留相应资源后发送的。
在此情形下，部署模块222向VIM请求分配所述VDU列表中各VDU所需 的VM资源时，具体用于经由所述VNFM将所述请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。
具体实现中，作为另一种实施例，部署模块222向VIM请求分配所述VDU列表中各VDU所需的VM资源时，具体用于依次经由VNFM和NFVO将请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。该实施例与上一实施例的不同在于，由NFVO直接向VIM发起资源分配请求，而上一实施例在发起资源分配请求之前还包括操作授权请求及响应确定网络中有相关资源。
作为一种可替代方案，具体实现中，如图24所示，本发明实施例的网元管理设备可包括：接收模块23、确定模块24以及扩容模块22，其中，扩容模块22为图21或图23中的扩容模块，在此不赘述，其中，
接收模块23，用于从VMFM接收VNF实例扩容请求；
确定模块24，用于根据所述接收模块23接收的VNF实例扩容请求和所述网元管理设备存储的资源配置信息，确定执行VNF实例扩容所需部署的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源。
其中，如图25所示，作为一种实施例，确定模块24可包括：VDU列表确定模块241、软件确定模块242、VM规格确定模块243以及链路确定模块244，其中：
VDU列表确定模块241，用于根据所述VNF实例扩容请求中携带的VNF实例标识及扩容后的容量规格，确定要部署的VDU列表；
软件确定模块242，用于根据所述网元管理设备存储的资源配置信息中的VDU与软件的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自对应的软件；
VM规格确定模块243，用于根据所述网元管理设备存储的资源配置信息中的VDU与虚拟机(Virtual Machine，VM)的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自所需的VM规格；
链路确定模块244，用于根据所述网元管理设备存储的资源配置信息中的网 络互通关系及各链路的带宽需求，所述VM各自对链路的需求。
在其他实施例中，除上述模块之外，确定模块24还可包括隔离亲和度确定模块245，用于根据所述网元管理设备存储的资源配置信息中的VDU隔离度和亲和度，确定VM的亲和度和隔离度。
其中，软件确定模块242、VM规格确定模块243、链路确定模块244以及隔离亲和度确定模块245分别与软件确定模块212、VM规格确定模块213、链路确定模块214以及隔离亲和度确定模块215相同，因此，在此，对它们的具体实现不再赘述。
由上可见，本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行VNF实例扩容的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了VNF实例扩容的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行VNF实例扩容时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来执行VNF实例扩容的具体流程也可不同。
图26为本发明实施例的网元管理设备的另一实施例的结构组成示意图。如图26所示，其可包括：获取模块30、确定模块31以及缩容模块32，其中：
获取模块30，用于获取VNF上报的性能参数；
具体实现中，本发明实施例获取的VNF上报的性能参数可包括CPU占用率，或者其他任意可通过VNF上报的性能参数。
确定模块31，用于根据所述获取模块30获取的上报的性能参数及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，所述要终止的VDU列表中包括一个或多个需终止的VDU，所述资源配置信息中的VNF实例缩容策略包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一 阈值与一缓冲参数的差值时，执行VNF实例缩容，终止一个或多个指定类型的VDU。
具体实现中，本发明实施例可在网元管理设备内存储的资源配置信息中可维护一份VNF实例扩缩容策略信息，或者所述资源配置信息中可分别维护一份VNF实例扩容策略信息及一份VNF实例缩容策略信息，其中所述VNF实例扩容策略可包括：当上一次上报的性能参数低于第二阈值，本次上报的性能参数大于或等于第二阈值时，执行VNF实例扩容，即在已有的VNF实例容量规格的基础上再部署一个或多个指定类型的VDU；所述VNF实例缩容策略可包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止VNF实例中一个或多个指定类型的VDU。
具体实现中，本实施例存储的VNF扩缩容策略及VNF上报的性能参数的举例可参考步骤S311的相关描述，在此不进行赘述。这样，当VNF上报的CPU的占用率符合上述4个策略中的第3个策略或第4个策略时，所述确定模块31可确定执行VNF实例缩容所需终止的VDU列表，比如，当满足上述策略3时，确定在所述VNF原有实例的基础上需终止一个类型为A的VDU。
具体实现中，本发明实施例在网元管理设备中存储的资源配置信息中还可包括容量规格与虚拟部署单元VDU的配比关系、VDU与软件的对应关系、VDU与虚拟机的对应关系、网络互通关系及各链路的带宽需求等信息。由此，本发明实施例，当确定模块31确定需终止的VDU列表之后，可进一步根据网元管理设备中存储的资源配置信息进一步确定所述VDU列表中各VDU所需释放的虚拟机VM资源及各VDU需删除的软件。其中，所述虚拟机VM资源可包括：各VDU所需的虚拟机VM规格、以及各VM对链路的需求等。进一步，作为补充实施例，本发明实施例在的确定模块31还可确定需终止的与所述确定的需终止的VDU具有亲和度和隔离度关系的VDU。
作为一种实施方式，如图27所示，本发明实施例的确定模块31具体可包括：VDU列表确定模块311和释放资源确定模块312，其中：
VDU列表确定模块311，用于根据所述获取模块30获取的VNF上报的性能参数及所述网元管理设备中存储的资源配置信息中的VNF实例缩容策略信息，确定执行VNF实例缩容所需终止的VDU列表。具体如何确定参考前面的描述，在此不进行赘述。
释放资源确定模块312，用于根据所述资源配置信息中各VDU的基础数据配置信息，确定所述需终止的VDU列表中各VDU需删除的软件、需释放的VM规格、以及需释放的与VM相关的链路。具体实现中，进行基础数据配置是指将为各VDU加载的软件及各VDU分配的虚拟机VM资源的分配结果及各VDU的类型等信息记录到所述网元管理设备中作为后续可用的资源配置信息，以便所述网元管理设备基于这些基础数据配置对VDU进行后续管理。因此，每个VDU的基础数据配置包括了VDU对对应的VM规格，VDU对应的软件，VDU对应的VM的链路等信息。
具体实现中，不同类型的VDU所对应的VM规格可不相同。VDU与VM规格的对应关系的示例可为步骤S3111所举示例，在此不进行赘述。
具体实现中，本实施例的网元管理设备内部存储的资源配置信息中网元的互通关系及带宽需求示例可如步骤S3111所示，在此不进行赘述。
因此，本发明实施例所述释放资源确定模块312还可根据上述互通关系，确定需释放VM_A与VM_B之间的链路，以及释放VM_B与VM_C之间的链路。
具体实现中，资源释放确定模块321在删除各VDU对应的软件、需释放的VM规格、以及需释放的与VM相关的链路之后，还可删除所述网元管理单元中配置的VDU的基础数据，比如，解除所述VDU与VNF实例的标识的关联关系，删除存储的所述VDU与软件的对应关系信息，删除所述VDU与VM规格的对应关系，以及删除所述VDU的VM的相关的链路信息等。
具体实现中，在其他实施例中，所述资源释放确定模块312还用于根据所述资源配置信息中的VM隔离度和亲和度，确定需终止的与VDU列表中VDU具有亲和度和隔离度关系的VDU。具体实现中，本实施例网元管理设备存储的 资源配置信息中的VM隔离度和亲和度的示例可为步骤S31112所示，在此不进行赘述。
因此，比如假设确定需终止VDU_A，在终止位于物理服务器1上的类型为VDU_A的编号为VDU_1的VDU时，还要终止位于物理服务器1上的类型为VDU_C的编号为VDU_3的VDU。
缩容模块32，用于根据所述确定模块31确定的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例缩容。
作为一种实施例，如图28所示，本发明实施例的缩容模块32具体包括：
生成模块321，用于根据所述确定模块31确定的要终止的VDU列表生成删除VDU的命令列表。作为一种实施例，确定模块31所确定的列表中包括1个类型为A的VDU(简化为：1*VDU_A)。
具体实现中，本发明实施例网元管理设备内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，生成模块321即可为确定的VDU列表中的各VDU分别生成删除命令，整个VDU列表中的各VDU生成的删除命令集合即形成VDU删除命令列表。作为一种示例，当确定的VDU列表为：1*VDU_A时，本发明实施例可生成如下的VDU删除命令列表：
RMV VDU：VDU_1(表示终止的VDU的编号)，VDU_A(表示终止的VDU类型)；
删除模块322，用于执行所述删除VDU的命令列表中的VDU删除命令，并在执行所述删除VDU的命令列表中的VDU删除命令的过程中，卸载各VDU的软件，并通过VNFM向所述VIM请求释放各VDU相关的虚拟机VM资源，并将所述网元管理设备中与所述VDU相关的基础数据删除。
通知模块323，用于在所述删除模块322执行完所述删除VDU的命令列表中的所有VDU删除命令后，通知VNFM所述VNF实例化缩容完成。
由上可见，本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行VNF实例缩容的生命周期管理时也不需 再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了VNF实例缩容的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行VNF实例缩容时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来执行VNF实例缩容的具体流程也可不同。另外，本发明实施例的VNF实例缩容策略中通过一缓冲参数来降低了VNF的乒乓弹性伸缩。
作为一种可替代的方案，如图29，具体实现中，本发明实施例的网元管理设备可包括：接收模块33、确定模34以及缩容模块32，其中，缩容模块32与图28和图26中的缩容模块相同，在此不进行赘述，其中：
接收模块33，用于从VNFM接收VNF实例缩容请求；
确定模块34，用于根据所述接收模块33接收的缩容VNF实例请求和所述网元管理设备存储的资源配置信息，确定执行VNF实例缩容所需终止的VDU列表及所述VDU列表中各VDU所对应的需释放的VM资源。
其中，如图30所示，作为一种实施例，确定模块34可包括：VDU列表确定模块341和释放资源确定模块342，其中：
VDU列表确定模块341，用于根据所述接收模块33接收的缩容VNF实例请求中携带的VNF实例标识及缩容后的容量规格，确定执行VNF实例缩容需终止的VDU列表；
释放资源确定模块342，用于根据所述资源配置信息中各VDU的基础数据配置信息，确定所述需终止的VDU列表中各VDU需删除的软件、需释放的VM规格、以及需释放的与VM相关的链路。
具体实现中，在其他实施例中，所述释放资源确定模块342还可用于根据所述资源配置信息中的VM隔离度和亲和度，确定需终止的与VDU列表中VDU具有亲和度和隔离度关系的VDU。
其中，本实施例的释放资源确定模块342与上一实施例的释放资源确定模块322相同，在此，对其的具体实现不进行赘述。
图31为本发明实施例的网元管理设备的另一实施例的结构组成示意图。如图31所述，其可包括：接收模块40、确定模块41以及终止模块42，其中：
接收模块40，用于接收VNF管理器VMFM发送的终止VNF实例请求消息。
确定模块41，用于根据所述接收模块40接收的终止VNF实例请求消息和所述网元管理设备中存储的资源配置信息确定执行终止VNF实例需要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源。
作为一种实施方式，如图32所示，确定模块41具体可包括：VDU列表确定模块411和释放资源确定模块412，其中：
VDU列表确定模块411，用于根据所述终止VNF实例请求消息中携带的VNF实例标识和所述网元管理设备中存储的资源配置信息，确定要终止的VNF实例及需终止的VDU列表；
释放资源确定模块412，用于根据所述资源配置信息中各VDU的基础数据配置信息，确定所述需终止的VDU列表中各VDU需删除的软件、需释放的VM规格、以及需释放的与VM相关的链路。具体实现中，进行基础数据配置是指将为各VDU加载的软件及各VDU分配的虚拟机VM资源的分配结果及各VDU的类型等信息记录到所述网元管理设备中作为后续可用的资源配置信息，以便所述网元管理设备基于这些基础数据配置对VDU进行后续管理。因此，每个VDU的基础数据配置包括了VDU对对应的VM规格，VDU对应的软件，VDU对应的VM的链路等信息。
具体实现中，不同类型的VDU所对应的VM规格可不相同。具体实现中，本实施例中VDU与VM规格的对应关系的示例可为步骤S4111所举示例，在此不进行赘述。
因此，本发明实施例释放资源确定模块412即可根据VDU与VM规格的对应关系，确定需删除的VM。比如，当确定需终止VDU列表为1*VDU_A+1*VDU_B+1*VDU_C，可确定需删除1个VM_A，删除该VM_A具体表现为需释放4核vCPU+8G内存+2G带宽；需删除1个VM_B，删除该VM_B具体表现为需释放4核vCPU+16G内存+2G带宽；需删除1个VM_C，删除该 VM_C具体表现为需释放2核vCPU+8G内存+2G带宽。
具体实现中，本实施例的网元管理设备内部存储的资源配置信息中网元的互通关系及带宽需求示例可如步骤S4111所示，在此不进行赘述。
因此，本发明实施例释放资源确定模块412即可根据上述互通关系，确定需释放VM_A与VM_B之间的链路，以及释放VM_B与VM_C之间的链路。具体实现中，在其他实施例中，释放资源确定模块412还可根据所述资源配置信息中的VM隔离度和亲和度，确定需终止的与VDU列表中VDU具有亲和度和隔离度关系的VDU。
具体实现中，本实施例网元管理设备存储的资源配置信息中的VM隔离度和亲和度的示例可为步骤S41112所示，在此不进行赘述。
终止模块42，用于根据所述确定模块41确定出的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例终止。
作为一种实施例，如图33所示，终止模块42具体包括：
生成模块421，所述网元管理设备根据确定的要终止的VDU列表生成删除VDU的命令列表。作为一种实施例，假设确定模块41确定的列表为1*VDU_A+1*VDU_B+1*VDU_C。
具体实现中，本发明实施例网元管理设备内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，生成模块421可为确定的VDU列表中的各VDU分别生成删除命令，整个VDU列表中的各VDU生成的删除命令集合即形成VDU删除命令列表。
具体实现中，本实施例中生成的VDU删除命令列表的示例可为步骤S4121所举示例，在此不进行赘述。删除模块422，用于执行所述删除VDU的命令列表中的VDU删除命令，并在执行所述删除VDU的命令列表中的VDU删除命令的过程中，卸载各VDU的软件，并通过VNFM向所述VIM请求释放各VDU相关的虚拟机VM资源，并将所述网元管理设备中与所述VDU相关的基础数据删除。
通知模块423，用于在所述删除模块422执行完所述删除VDU的命令列表 中的所有VDU删除命令后，通知VNFM所述VNF实例化终止完成。
由上可见，本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行VNF实例终止的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了VNF实例终止的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行VNF实例终止时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来执行VNF实例终止的具体流程也可不同。
以上列举了网元管理设备的功能模块组成实施例，下面对网元管理设备的硬件组成及其各部分硬件的分工协作进行举例说明。
图34为本发明实施例的网元管理设备的另一实施例的结构组成示意图。如图34所示，其可包括：输入装置50、输出装置51、通信链路52、收发装置53、存储器54以及处理器55，其中：
所述输入装置50，用于接收外部输入所述网元管理设备的输入数据。具体实现中，本发明实施例所述的输入装置50可包括键盘、鼠标、光电输入装置、声音输入装置、触摸式输入装置、扫描仪等。所述输入数据可包括各种用户指令。
所述输出设备51，用于对外输出所述网元管理设备的输出数据。具体实现中，本发明实施例所述的输出装置51可包括显示器、扬声器、打印机等。所述输出数据可包括对用户指令的响应数据。
所述通信链路52，用于与其他设备进行通讯。具体实现中，本发明实施例所述的通信链路52可是传播介质的一个实例。传播介质一般可以将计算机可读指令、数据结构、程序模块或其他调制数据信号(诸如载波或其他传送机制)形式的其他数据具体化，举例来说，传播介质可包括有线媒体、诸如优先网络或直线连接，传播介质还可包括无线介质，比如声波、射频、红外线等。具体 实现中，本发明实施例的通信链路52可用于传输VNF实例化请求消息以及传输向VNFM发送的实例化VNF完成的消息。
所述收发装置53，用于通过所述通信链路52向其他设备发送数据或者从其他设备接收数据。本发明实施例的收发装置53可用于接收VNF实例化请求消息以及向VNFM发送实例化VNF完成的消息。具体实现中，所述收发装置53可为与VNFM之间的Ve-Vnfm接口。
所述存储器54，用于存储带有各种功能的程序数据。本发明实施例中存储器54存储的数据包括资源配置信息，以及被处理器调用并运行的程序数据。具体实现中，本发明实施例的存储器54可以是系统存储器，比如，挥发性的(诸如RAM)，非易失性的(诸如ROM，闪存等)，或者两者的结合。具体实现中，本发明实施例的存储器54还可以是系统之外的外部存储器，比如，磁盘、光盘、磁带等。
所述处理器55，用于调用所述存储器54中存储的程序数据和资源配置信息，并执行如下操作：
A1：从所述收发装置53接收VNF管理器VNFM发送的实例化VNF请求消息。
A2：根据所述实例化VNF请求消息和所述存储器54中存储的资源配置信息确定执行实例化VNF所需的资源。
A3：根据所述确定出的资源来实例化VNF。
具体实现中，本发明实施例的所述网元管理设备与所述VNF由同一供应商提供，可包括但不限于网元管理系统(Element Management System，EMS)或VNF本地管理单元(Local Management Unit，LMU)。
具体实现中，当接收到实例化VNF请求消息之后，处理器55可在网元管理设备中为所述实例化VNF请求消息所请求建立的VNF实例创建实例标识，并建立所述VNF实例的版本号、类型及容量规格与所述实例标识的对应关系。具体实现中，处理器55可根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定实例化VNF所需的虚拟部署单元VDU列表，以及 所述VDU列表中各VDU所需虚拟机VM资源。其中，所述虚拟机VM资源可包括：各VDU所需的虚拟机VM规格、以及各VM对链路的需求等。进一步，作为补充实施例，本发明实施例所确定的执行实例化VNF所需的资源还可包括：各VDU对应的软件及各VM的亲和度和隔离度。
具体实现中，处理器55根据所述实例化VNF请求消息和所述存储器54中存储的资源配置信息确定执行实例化VNF所需的资源时，具体执行如下操作：
A21，根据所述实例化VNF的请求确定要部署的VNF的类型、版本号及容量规格。
具体实现中，处理器55可根据所述实例化VNF请求中携带的VNF的信息模型描述模版VNFD索引信息和容量规格，到对应的VNFD中查询要部署的VNF的类型，版本号及容量规格。如何通过VNFD索引找到对应的VNFD及查询到要部署的VNF的类型，版本号，容量规格的示例详见方法步骤S1110，在此不进行赘述。
或者，处理器55可根据所述实例化VNF请求携带的VNF的类型、版本号及容量规格确定要部署的VNF的类型、版本号及容量规格。如何根据请求携带的VNF的类型、版本号及容量规格确定要部署的VNF的类型、版本号及容量规格的示例，详见方法流程S1110，在此不进行赘述。
具体实现中，所述实例化VNF请求携带的VNF的类型、版本号及容量规格注册在所述VNFD中或注册在网络业务的信息模型描述模版NSD中。
A22，根据所述资源配置信息中的容量规格与虚拟部署单元VDU的配比关系，确定所述要部署的VNF所需的VDU列表。
本发明实施例中，处理器55确定所要部署的VNF所需的VDU列表的具体手段可与步骤S1111相同，在此不进行赘述。
A23，根据所述资源配置信息中的VDU与软件的对应关系，确定所述VDU列表中各VDU各自对应的软件。
A24，根据所述资源配置信息中的VDU与虚拟机(Virtual Machine，VM)的对应关系，确定各VDU各自所需的VM规格。具体实现中，不同类型的VDU 所需的VM规格可不相同。关于VDU所需的VM规格的示例可为步骤S1113所举示例，在此不进行赘述。A25，根据所述资源配置信息中的网络互通关系及各链路的带宽需求，确定各VM各自对链路的需求。
比如，存储器54存储的资源配置信息中网元的互通关系及带宽需求可参考步骤S1114所示示例，在此不进行赘述，
则处理器55确定的链路需求可为VM_A与VM_B之间需建立虚拟链路，VM_B与VM_C之间需建立虚拟链路。
具体实现中，在其他实施例中，处理器55还根据所述资源配置信息中的VDU隔离度和亲和度，确定VM的亲和度和隔离度。具体实现中，存储器54存储的资源配置信息中的VDU隔离度和亲和度信息可与步骤S1115所举示例相同，在此不进行赘述。
因此，处理器55可根据网元管理设备存储的VDU隔离度和亲和度确定各VM的亲和度和隔离度为VM_A与VM_C部署在相同的物理服务器上，多个VM_A部署在不同的物理服务器上，比如，VM_1、VM_2的类型均是VM_A，但是在部署时，VM_1和VM_2需部署在不同的物理服务器上。
具体实现中，处理器55执行根据所述确定出的资源来实例化VNF时，具体可执行如下步骤：
A31，根据所述确定的VDU列表生成部署VDU的命令列表。
具体实现中，本发明实施例的存储器54存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，处理器55可为确定的VDU列表中的各VDU分别生成部署命令，整个VDU列表中的各VDU生成的部署命令集合即形成VDU部署命令列表。本实施例可生成的VDU部署命令列表示例可为步骤S1120所举示例，在此不进行赘述。A32，在执行所述命令列表中的VDU的部署命令过程中，向VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置。
具体实现中，处理器55可根据步骤B所确定的VM资源及链路资源向VIM 请求分配对应的资源，以及为所述VDU列表中的各VDU加载软件及进行基础数据配置。具体实现中，所述进行基础数据配置是指将为各VDU加载的软件及各VDU分配的虚拟机VM资源的分配结果及各VDU的类型等信息记录到所述网元管理设备中，并将各VDU与包括这些VDU的VNF实例的标识进行关联，作为后续可用的资源配置信息，以便所述网元管理设备基于这些基础数据配置对VNF实例及实例中的VDU进行管理。由此可见，本发明实施例在执行VNF实例化的过程中即自动完成基础数据配置，解决了现有技术需要手动在网元管理设备中进行基础数据配置的技术问题。
作为一种实施例，所述处理器55可逐一执行所述命令列表中的每一个VDU的部署命令，在执行每一个VDU的部署命令过程中，向所述VIM请求分配所述每一个VDU所需的虚拟机VM资源，并根据所述VIM为所述每一个VDU分配的资源为所述一个VDU加载软件并根据所述VIM对每一个VDU分配所需的虚拟机VM资源的分配结果，为所述每一个VDU加载软件并进行基础数据配置。
当然作为其他的实施方式，处理器55可一次性向VIM请求所述VDU列表中所有VDU所需的VM资源及链路资源，并在请求完之后，一次性为各VDU加载软件及进行基础数据配置。
A33，所述网元管理设备在所述VDU列表中所有VDU的部署命令执行完之后，通知所述VNFM所述实例化VNF完成。
具体实现中，作为一种实施例，本发明实施例处理器55在，还执行根据所述确定的VDU列表生成部署VDU的命令列表的步骤之前，还执行如下步骤：
经由VNF管理器(Virtual Network Function Manager，VNFM)向网络功能虚拟化编排器(Network Functions Virtualization Orchestrator，NFVO)发起实例化VNF的操作授权请求，并经所述VNFM从所述NFVO接收所述实例化VNF的操作授权的响应，所述实例化VNF的操作授权的响应是所述NFVO确定存在执行所述实例化VNF所需的资源并请求所述VIM预留相应资源后发送的；
在此情形下，处理器55向VIM请求分配所述VDU列表中各VDU所需的VM资源及链路资源，具体执行如下步骤：
经由所述VNFM将所述请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。
具体实现中，作为另一种实施例，处理器55向VIM请求分配所述VDU列表中各VDU所需的VM资源，具体执行如下步骤：
依次经由VNFM和NFVO将请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。该实施例与上一实施例的不同在于，由NFVO直接向VIM发起资源分配请求，而上一实施例在发起资源分配请求之前还包括操作授权请求及响应确定网络中有相关资源。
由上可见，本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行实例化VNF的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了实例化VNF的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行实例化VNF时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来实例化VNF的具体流程也可不同。
图35为本发明实施例的网元管理设备的另一实施例的结构组成示意图。如图35所示，其可包括：输入装置60、输出装置61、通信链路62、收发装置63、存储器64以及处理器65，其中：
所述输入装置60，用于接收外部输入所述网元管理设备的输入数据。具体实现中，本发明实施例所述的输入装置60可包括键盘、鼠标、光电输入装置、声音输入装置、触摸式输入装置、扫描仪等。所述输入数据可包括各种用户指令。
所述输出设备61，用于对外输出所述网元管理设备的输出数据。具体实现中，本发明实施例所述的输出装置61可包括显示器、扬声器、打印机等。所述输出数据可包括对用户指令的响应数据。
所述通信链路62，用于与其他设备进行通讯。具体实现中，本发明实施例所述的通信链路62可是传播介质的一个实例。传播介质一般可以将计算机可读指令、数据结构、程序模块或其他调制数据信号(诸如载波或其他传送机制)形式的其他数据具体化，举例来说，传播介质可包括有线媒体、诸如优先网络或直线连接，传播介质还可包括无线介质，比如声波、射频、红外线等。具体实现中，本发明实施例的通信链路62可用于传输VNF上报的性能参数以及传输向VNFM发送的VNF扩容完成的消息，或用于传输VNF扩容请求消息。
所述收发装置63，用于通过所述通信链路62向其他设备发送数据或者从其他设备接收数据。本发明实施例的收发装置63可用于接收VNF上报的性能参数，VNF实例扩容请求消息以及向VNFM发送VNF实例扩容完成的消息。具体实现中，所述收发装置63可为与VNFM之间的Ve-Vnfm接口。
所述存储器64，用于存储带有各种功能的程序数据。本发明实施例中存储器64存储的数据包括资源配置信息，以及被处理器调用并运行的程序数据。具体实现中，本发明实施例的存储器64可以是系统存储器，比如，挥发性的(诸如RAM)，非易失性的(诸如ROM，闪存等)，或者两者的结合。具体实现中，本发明实施例的存储器64还可以是系统之外的外部存储器，比如，磁盘、光盘、磁带等。
所述处理器65，用于调用所述存储器64中存储的程序数据和资源配置信息，并执行如下操作：
B1：获取VNF上报的性能参数或从VMFM接收VNF实例扩容请求。
具体实现中，本发明实施例获取的VNF上报的性能参数可包括CPU占用率，或者其他任意可通过VNF上报的性能参数。
B2：根据所述上报的性能参数及所述网元管理设备中存储的资源配置信息，确定执行VNF实例扩容所需部署的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源；或者，根据所述接收的VNF实例扩容请求和所述网元管理设备存储的资源配置信息，确定执行VNF实例扩容所需部署的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源。
B3：根据所述确定的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源，执行VNF实例扩容。
具体实现中，本发明实施例存储器64内存储的资源配置信息中可维护一份VNF实例扩缩容策略信息，或者所述资源配置信息中可分别维护一份VNF实例扩容策略信息及一份VNF实例缩容策略信息，其中所述VNF实例扩容策略可包括：当上一次上报的性能参数低于第二阈值，本次上报的性能参数大于或等于第二阈值时，执行VNF实例扩容，即在已有的VNF实例容量规格的基础上再部署一个或多个指定类型的VDU；所述VNF实例缩容策略可包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止VNF实例中一个或多个指定类型的VDU。
具体实现中，本实施例存储的VNF扩缩容策略及VNF上报的性能参数的举例可参考步骤S211的相关描述，在此不进行赘述。这样，当VNF上报的CPU的占用率符合上述4个策略中的第1个策略或第2个策略时，处理器65可确定执行VNF实例扩容所需部署的VDU列表，比如，当满足上述策略1时，在所述VNF原有实例的基础上确定需再部署一个类型为A的VDU。
具体实现中，本发明实施例在存储器64中存储的资源配置信息中还可包括容量规格与虚拟部署单元VDU的配比关系、VDU与软件的对应关系、VDU与虚拟机的对应关系、或网络互通关系及各链路的带宽需求等信息。由此，本发明实施例，当确定需部署的VDU列表之后，处理器65可进一步根据网元管理设备中存储的资源配置信息进一步确定所述VDU列表中各VDU所需的虚拟机VM资源。其中，所述虚拟机VM资源可包括：各VDU所需的虚拟机VM规格、以及各VM对链路的需求等。进一步，作为补充实施例，本发明实施例所确定的资源还可包括：各VDU对应的软件及各VM的亲和度和隔离度。
作为一种实施方式，处理器65在执行根据所述上报的性能参数及所述网元管理设备中存储的资源配置信息，确定执行VNF实例扩容所需部署的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源的步骤时，具体可执行如下步骤：
B21，根据所述上报的性能参数及所述网元管理设备中存储的资源配置信息中的VNF实例扩容策略信息，确定执行VNF实例扩容所需部署的VDU列表。具体如何确定VDU列表参考前面的描述，在此不进行赘述。
B22，根据所述资源配置信息中的VDU与软件的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自对应的软件。
B23，根据所述资源配置信息中的VDU与虚拟机(Virtual Machine，VM)的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自所需的VM规格。具体实现中，不同类型的VDU所需的VM规格可不相同。VDU与VM规格的对应关系举例可见步骤S2112的相关描述，在此不进行赘述。B24，根据所述资源配置信息中的网络互通关系及各链路的带宽需求，所述VM各自对链路的需求。
比如，存储器64内部存储的资源配置信息中网元的互通关系及带宽需求的示例可如步骤S2113所示，在此不进行赘述。
具体实现中，在其他实施例中，处理器65根据所述资源配置信息中的VDU隔离度和亲和度，确定VM的亲和度和隔离度。具体实现中，存储器64存储的资源配置信息中的VDU隔离度和亲和度的示例可如步骤S2114所示，在此不进行赘述。
因此，处理器65即可根据网元管理设备存储的VDU隔离度和亲和度确定各VM的亲和度和隔离度为VM_A与VM_C部署在相同的物理服务器上，多个VM_A部署在不同的物理服务器上，比如，VM_1、VM_2的类型均是VM_A，但是在部署时，VM_1和VM_2需部署在不同的物理服务器上。
作为一种可替代方案，具体实现中，处理器65在执行根据所述接收的VNF实例扩容请求和所述网元管理设备存储的资源配置信息，确定执行VNF实例扩容所需部署的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源时，具体执行如下步骤：
B25，根据所述VNF实例扩容请求中携带的VNF实例标识及扩容后的容量规格，确定要部署的VDU列表；
B26，根据所述资源配置信息中的VDU与软件的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自对应的软件；
B27，根据所述资源配置信息中的VDU与虚拟机(Virtual Machine，VM)的对应关系，确定执行VNF实例扩容所需部署的VDU列表中各VDU各自所需的VM规格；
B28，根据所述资源配置信息中的网络互通关系及各链路的带宽需求，所述VM各自对链路的需求。
具体实现中，所述处理器65执行根据所述确定的VDU列表及所述VDU列表中各VDU所需的虚拟机VM资源，执行VNF实例扩容的步骤时，具体执行如下步骤：
B31，根据所述确定的VDU列表生成部署VDU的命令列表。假设确定的VDU列表中包括1个类型为A的VDU(简化为：1*VDU_A)。
具体实现中，本发明实施例存储器64内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，处理器65可为确定的VDU列表中的各VDU分别生成部署命令，整个VDU列表中的各VDU生成的部署命令集合即形成VDU部署命令列表。作为一种示例，当确定的VDU列表为：1*VDU_A时，本发明实施例可生成如下的VDU部署命令列表：
ADD VDU：VDU_1(表示部署的VDU的编号)，VDU_A(表示部署的VDU类型)；
B32，在执行所述命令列表中的VDU的部署命令过程中，向VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置。
具体实现中，所述进行基础数据配置是指将为各VDU加载的软件及各VDU分配的虚拟机VM资源的分配结果及各VDU的类型等信息记录到所述网元管理设备中，并将各VDU与包括这些VDU的VNF实例的标识进行关联，作为后续可用的资源配置信息，以便所述网元管理设备基于这些基础数据配置对VNF实 例及实例中的VDU进行管理。由此可见，本发明实施例在执行VNF实例扩容的过程中即自动完成基础数据配置，解决了现有技术需要手动在网元管理设备中进行基础数据配置的技术问题。
作为一种实施例，处理器65可逐一执行所述命令列表中的每一个VDU的部署命令，在执行每一个VDU的部署命令过程中，向所述VIM请求分配所述每一个VDU所需的虚拟机VM资源，并根据所述VIM为所述每一个VDU分配的资源为所述一个VDU加载软件并根据所述VIM对每一个VDU分配所需的虚拟机VM资源的分配结果，为所述每一个VDU加载软件并进行基础数据配置。
当然作为其他的实施方式，处理器65可一次性向VIM请求所述VDU列表中所有VDU所需的VM资源及链路资源，并在请求完之后，一次性为各VDU加载软件及进行基础数据配置。
B33，在所述VDU列表中所有VDU的部署命令执行完之后，通知所述VNFM所述VNF实例扩容完成。
具体实现中，作为一种实施例，处理器65在执行根据所述确定的VDU列表生成部署VDU的命令列表的步骤之前，还执行如下步骤：
经由VNF管理器(Virtual Network Function Manager，VNFM)向网络功能虚拟化编排器(Network Functions Virtualization Orchestrator，NFVO)发起实例化VNF的操作授权请求，并经所述VNFM从所述NFVO接收所述实例化VNF的操作授权的响应，所述实例化VNF的操作授权的响应是所述NFVO确定存在执行所述实例化VNF所需的资源并请求所述VIM预留相应资源后发送的；
在此情形下，处理器65向VIM请求分配所述VDU列表中各VDU所需的VM资源时，具体用于经由所述VNFM将所述请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。
具体实现中，作为另一种实施例，处理器65向VIM请求分配所述VDU列表中各VDU所需的VM资源时，具体用于依次经由VNFM和NFVO将请求分配所述VDU列表中各VDU所需的VM资源的请求转发给所述VIM。该实施例与上一实施例的不同在于，由NFVO直接向VIM发起资源分配请求，而上一实 施例在发起资源分配请求之前还包括操作授权请求及响应确定网络中有相关资源。
由上可见，本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行VNF实例扩容的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了VNF实例扩容的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行VNF实例扩容时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来执行VNF实例扩容的具体流程也可不同。
图36为本发明实施例的网元管理设备的另一实施例的结构组成示意图。如图36所示，其可包括：输入装置70、输出装置71、通信链路72、收发装置73、存储器74以及处理器75，其中：
所述输入装置70，用于接收外部输入所述网元管理设备的输入数据。具体实现中，本发明实施例所述的输入装置70可包括键盘、鼠标、光电输入装置、声音输入装置、触摸式输入装置、扫描仪等。所述输入数据可包括各种用户指令。
所述输出设备71，用于对外输出所述网元管理设备的输出数据。具体实现中，本发明实施例所述的输出装置71可包括显示器、扬声器、打印机等。所述输出数据可包括对用户指令的响应数据。
所述通信链路72，用于与其他设备进行通讯。具体实现中，本发明实施例所述的通信链路72可是传播介质的一个实例。传播介质一般可以将计算机可读指令、数据结构、程序模块或其他调制数据信号(诸如载波或其他传送机制)形式的其他数据具体化，举例来说，传播介质可包括有线媒体、诸如优先网络或直线连接，传播介质还可包括无线介质，比如声波、射频、红外线等。具体实现中，本发明实施例的通信链路72可用于传输VNF上报的性能参数以及传 输向VNFM发送的VNF缩容完成的消息，或用于传输VNF缩容请求消息。
所述收发装置73，用于通过所述通信链路72向其他设备发送数据或者从其他设备接收数据。本发明实施例的收发装置73可用于接收VNF上报的性能参数，VNF实例缩容请求消息以及向VNFM发送VNF实例缩容完成的消息。具体实现中，所述收发装置73可为与VNFM之间的Ve-Vnfm接口。
所述存储器74，用于存储带有各种功能的程序数据。本发明实施例中存储器74存储的数据包括资源配置信息，以及被处理器调用并运行的程序数据。具体实现中，本发明实施例的存储器74可以是系统存储器，比如，挥发性的(诸如RAM)，非易失性的(诸如ROM，闪存等)，或者两者的结合。具体实现中，本发明实施例的存储器74还可以是系统之外的外部存储器，比如，磁盘、光盘、磁带等。
所述处理器75，用于调用所述存储器74中存储的程序数据和资源配置信息，并执行如下操作：
C1：获取VNF上报的性能参数或接收VNF实例缩容请求；
具体实现中，本发明实施例获取的VNF上报的性能参数可包括CPU占用率，或者其他任意可通过VNF上报的性能参数。
C2：根据所述上报的性能参数及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，所述要终止的VDU列表中包括一个或多个需终止的VDU，所述资源配置信息中的VNF实例缩容策略包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止一个或多个指定类型的VDU；或者，根据所述接收的VNF实例缩容请求及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源。
C3：根据所述确定的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例缩容。
具体实现中，本发明实施例可在存储器74内存储的资源配置信息中可维护 一份VNF实例扩缩容策略信息，或者所述资源配置信息中可分别维护一份VNF实例扩容策略信息及一份VNF实例缩容策略信息，其中所述VNF实例扩容策略可包括：当上一次上报的性能参数低于第二阈值，本次上报的性能参数大于或等于第二阈值时，执行VNF实例扩容，即在已有的VNF实例容量规格的基础上再部署一个或多个指定类型的VDU；所述VNF实例缩容策略可包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止VNF实例中一个或多个指定类型的VDU。
具体实现中，本实施例存储的VNF扩缩容策略及VNF上报的性能参数的举例可参考步骤S311的相关描述，在此不进行赘述。这样，当VNF上报的CPU的占用率符合上述4个策略中的第3个策略或第4个策略时，处理器75可确定执行VNF实例缩容所需终止的VDU列表，比如，当满足上述策略3时，确定在所述VNF原有实例的基础上需终止一个类型为A的VDU。
具体实现中，本发明实施例在存储器74中存储的资源配置信息中还可包括容量规格与虚拟部署单元VDU的配比关系、VDU与软件的对应关系、VDU与虚拟机的对应关系、网络互通关系及各链路的带宽需求等信息。由此，本发明实施例，当确定需终止的VDU列表之后，处理器75可进一步根据网元管理设备中存储的资源配置信息进一步确定所述VDU列表中各VDU所需释放的虚拟机VM资源及各VDU需删除的软件。其中，所述虚拟机VM资源可包括：各VDU所需的虚拟机VM规格、以及各VM对链路的需求等。进一步，作为补充实施例，本发明实施例处理器75还可确定需终止的与所述确定的需终止的VDU具有亲和度和隔离度关系的VDU。
作为一种实施方式，本发明实施例处理器75在根据所述上报的性能参数及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源时，具体可执行如下步骤：
C21，根据所述上报的性能参数及所述网元管理设备中存储的资源配置信息 中的VNF实例缩容策略信息，确定执行VNF实例缩容所需终止的VDU列表。具体如何确定参考前面的描述，在此不进行赘述。
C22，根据所述资源配置信息中各VDU的基础数据配置信息，确定所述需终止的VDU列表中各VDU需删除的软件、需释放的VM规格、以及需释放的与VM相关的链路。
具体实现中，不同类型的VDU所对应的VM规格可不相同。
VDU与VM规格的对应关系的示例可为步骤S3111所举示例，在此不进行赘述。
具体实现中，本实施例的网元管理设备内部存储的资源配置信息中网元的互通关系及带宽需求示例可如步骤S3111所示，在此不进行赘述。因此，本发明实施例处理器75即可根据上述互通关系，确定需释放VM_A与VM_B之间的链路，以及释放VM_B与VM_C之间的链路。
具体实现中，处理器65在删除各VDU对应的软件、需释放的VM规格、以及需释放的与VM相关的链路之后，还可删除所述网元管理单元中配置的VDU的基础数据，比如，解除所述VDU与VNF实例的标识的关联关系，删除存储的所述VDU与软件的对应关系信息，删除所述VDU与VM规格的对应关系，以及删除所述VDU的VM的相关的链路信息等。
具体实现中，在其他实施例中，处理器75还可根据所述资源配置信息中的VM隔离度和亲和度，确定需终止的与VDU列表中VDU具有亲和度和隔离度关系的VDU。具体实现中，本实施例存储器74存储的资源配置信息中的VM隔离度和亲和度的示例可为步骤S31112所示，在此不进行赘述因此，比如假设确定需终止VDU_A，在终止位于物理服务器1上的类型为VDU_A的编号为VDU_1的VDU时，还要终止位于物理服务器1上的类型为VDU_C的编号为VDU_3的VDU。
作为一种可替代的方案，本发明实施例处理器75在执行根据所述接收的VNF实例缩容请求及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM 资源的步骤时，具体可执行如下步骤：
C23，从VNFM接收缩容VNF实例请求；
C24，根据所述接收的缩容VNF实例请求和所述网元管理设备存储的资源配置信息，确定执行VNF实例缩容所需终止的VDU列表及所述VDU列表中各VDU所对应的需释放的VM资源。
其中，作为一种实施例，处理器75根据所述接收的缩容VNF实例请求和所述网元管理设备存储的资源配置信息，确定执行VNF实例缩容所需终止的VDU列表及所述VDU列表中各VDU所对应的需释放的VM资源时，具体可执行如下步骤：
C241，根据所述接收的缩容VNF实例请求中携带的VNF实例标识及缩容后的容量规格，确定执行VNF实例缩容需终止的VDU列表；
C242，根据所述资源配置信息中各VDU的基础数据配置信息，确定所述需终止的VDU列表中各VDU需删除的软件、需释放的VM规格、以及需释放的与VM相关的链路。
具体实现中，在其他实施例中，处理器75还可根据所述资源配置信息中的VM隔离度和亲和度，确定需终止的与VDU列表中VDU具有亲和度和隔离度关系的VDU。
具体实现中，处理器75执行根据所述确定的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例缩容的步骤时，具体执行如下步骤：
C31，根据确定的要终止的VDU列表生成删除VDU的命令列表。假设确定的列表中包括1个类型为A的VDU(简化为：1*VDU_A)。
具体实现中，本发明实施例网元管理设备内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，处理器75可为确定的VDU列表中的各VDU分别生成删除命令，整个VDU列表中的各VDU生成的删除命令集合即形成VDU删除命令列表。作为一种示例，当确定的VDU列表为：1*VDU_A时，本发明实施例可生成如下的VDU删除命令列表：
RMV VDU：VDU_1(表示终止的VDU的编号)，VDU_A(表示终止的VDU类型)；
C32，在执行所述删除VDU的命令列表中的VDU删除命令的过程中，卸载各VDU的软件，并通过VNFM向所述VIM请求释放各VDU相关的虚拟机VM资源，并将所述网元管理设备中与所述VDU相关的基础数据删除。
C33，在执行完所述删除VDU的命令列表中的所有VDU删除命令后，通知VNFM所述VNF实例化缩容完成。
由上可见，本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行VNF实例缩容的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了VNF实例缩容的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行VNF实例缩容时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来执行VNF实例缩容的具体流程也可不同。另外，本发明实施例的VNF实例缩容策略中通过一缓冲参数来降低了VNF的乒乓弹性伸缩。
图37为本发明实施例的网元管理设备的另一实施例的结构组成示意图。如图37所示，其可包括：输入装置80、输出装置81、通信链路82、收发装置83、存储器84以及处理器85，其中：
所述输入装置80，用于接收外部输入所述网元管理设备的输入数据。具体实现中，本发明实施例所述的输入装置80可包括键盘、鼠标、光电输入装置、声音输入装置、触摸式输入装置、扫描仪等。所述输入数据可包括各种用户指令。
所述输出设备81，用于对外输出所述网元管理设备的输出数据。具体实现中，本发明实施例所述的输出装置81可包括显示器、扬声器、打印机等。所述输出数据可包括对用户指令的响应数据。
所述通信链路82，用于与其他设备进行通讯。具体实现中，本发明实施例所述的通信链路82可是传播介质的一个实例。传播介质一般可以将计算机可读指令、数据结构、程序模块或其他调制数据信号(诸如载波或其他传送机制)形式的其他数据具体化，举例来说，传播介质可包括有线媒体、诸如优先网络或直线连接，传播介质还可包括无线介质，比如声波、射频、红外线等。具体实现中，本发明实施例的通信链路82可用于传输终止VNF实例请求消息以及传输向VNFM发送的VNF实例终止完成的消息。
所述收发装置83，用于通过所述通信链路82向其他设备发送数据或者从其他设备接收数据。本发明实施例的收发装置83可用于接收终止VNF实例请求消息以及向VNFM发送VNF实例终止完成的消息。具体实现中，所述收发装置83可为与VNFM之间的Ve-Vnfm接口。
所述存储器84，用于存储带有各种功能的程序数据。本发明实施例中存储器84存储的数据包括资源配置信息，以及被处理器调用并运行的程序数据。具体实现中，本发明实施例的存储器84可以是系统存储器，比如，挥发性的(诸如RAM)，非易失性的(诸如ROM，闪存等)，或者两者的结合。具体实现中，本发明实施例的存储器84还可以是系统之外的外部存储器，比如，磁盘、光盘、磁带等。
所述处理器85，用于调用所述存储器84中存储的程序数据和资源配置信息，并执行如下操作：
D1：接收VNF管理器VMFM发送的终止VNF实例请求消息。
D2：根据所述终止VNF实例请求消息和所述网元管理设备中存储的资源配置信息确定执行终止VNF实例需要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源。
D3：根据所述确定出的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例终止。
作为一种实施方式，本发明实施例处理器85在执行根据所述终止VNF实例请求消息和所述网元管理设备中存储的资源配置信息确定执行终止VNF实例 需要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源的步骤时，具体执行如下步骤：
D21，根据所述终止VNF实例请求消息中携带的VNF实例标识和所述网元管理设备中存储的资源配置信息，确定要终止的VNF实例及需终止的VDU列表；
D22，所述资源配置信息中各VDU的基础数据配置信息，确定所述需终止的VDU列表中各VDU需删除的软件、需释放的VM规格、以及需释放的与VM相关的链路。
具体实现中，不同类型的VDU所对应的VM规格可不相同。
具体实现中，本实施例中VDU与VM规格的对应关系的示例可为步骤S4111所举示例，在此不进行赘述。
因此，本发明实施例处理器85即可根据VDU与VM规格的对应关系，确定需删除的VM。比如，当确定需终止VDU列表为1*VDU_A+1*VDU_B+1*VDU_C，可确定需删除1个VM_A，删除该VM_A具体表现为需释放4核vCPU+8G内存+2G带宽；需删除1个VM_B，删除该VM_B具体表现为需释放4核vCPU+16G内存+2G带宽；需删除1个VM_C，删除该VM_C具体表现为需释放2核vCPU+8G内存+2G带宽。
具体实现中，本实施例的存储器84内部存储的资源配置信息中网元的互通关系及带宽需求示例可如步骤S4111所示，在此不进行赘述。
因此，本发明实施例处理器85即可根据上述互通关系，确定需释放VM_A与VM_B之间的链路，以及释放VM_B与VM_C之间的链路。具体实现中，在其他实施例中，处理器85还可根据所述资源配置信息中的VM隔离度和亲和度，确定需终止的与VDU列表中VDU具有亲和度和隔离度关系的VDU。
具体实现中，本实施例存储器84存储的资源配置信息中的VM隔离度和亲和度的示例可为步骤S41112所示，在此不进行赘述。
因此，比如假设确定需终止VDU_A，在终止位于物理服务器1上的类型为VDU_A的编号为VDU_1的VDU时，还要终止位于物理服务器1上的类型为 VDU_C的编号为VDU_3的VDU。
具体实现中，本发明实施例处理器85执行根据所述确定出的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例终止的步骤时，具体执行如下步骤：
D31，根据确定的要终止的VDU列表生成删除VDU的命令列表。假设确定的列表为1*VDU_A+1*VDU_B+1*VDU_C。
具体实现中，本发明实施例存储器84内存储了各类型的配置命令，比如，部署命令ADD VDU；删除命令RMV VDU等。因此，处理器85可为确定的VDU列表中的各VDU分别生成删除命令，整个VDU列表中的各VDU生成的删除命令集合即形成VDU删除命令列表。
具体实现中，本实施例中生成的VDU删除命令列表的示例可为步骤S4121所举示例，在此不进行赘述。D32，在执行所述删除VDU的命令列表中的VDU删除命令的过程中，卸载各VDU的软件，并通过VNFM向所述VIM请求释放各VDU相关的虚拟机VM资源，并将所述网元管理设备中与所述VDU相关的基础数据删除。
D33，在执行完所述删除VDU的命令列表中的所有VDU删除命令后，通知VNFM所述VNF实例化终止完成。
由上可见，本发明实施例将VNFM的生命周期管理工作转移到网元管理设备之后，由此，VNFD中可无需再设置完整的能覆盖各个VNF生命周期管理过程的参数或脚本。而网元管理设备执行VNF实例终止的生命周期管理时也不需再受到VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，避免了VNF实例终止的生命周期管理过程过于机械化和缺乏灵活性。另外，网元管理设备与VNF通常属于同一供应商，因此，本发明实施例能支持各供应商执行VNF实例终止时的差异性，比如，不同供应商的网元管理设备中存储的资源配置信息的格式就可以不相同，以及不同供应商的网元管理设备根据确定出的资源来执行VNF实例终止的具体流程也可不同。
以上对本发明实施例的VNF生命周期管理方法及网元管理设备的实施例进 行了列举，下面结合具体的应用场景，对本发明实施例的VNF实例化过程、VNF扩容/缩容过程以及VNF实例化终止过程的详细流程进行说明。
图38为本发明的VNF实例化过程的原理流程示意图。如图38所示，其可包括：
S01，VNFM收到实例化VNF请求，请求方可以为NFVO等；请求中携带的VNF的信息模型描述模版VNFD索引信息和容量规格，到对应的VNFD中查询要部署的VNF的类型，版本号及容量规格。或者，请求携带的VNF的类型、版本号及容量规格确定要部署的VNF的类型、版本号及容量规格。
S02，VNFM根据Vendor/VNF类型等信息确定EMS(本例所使用的网元管理设备)，将请求转发到EMS。
S03，EMS根据请求中携带的信息及存储的资源配置信息，确定执行实例化VNF所需的资源。
S04，EMS携带执行实例化VNF所需的资源，发起实例化VNF操作的授权请求给VNFM。
S05，VNFM将实例化VNF操作的授权请求转发到NFVO。
S06，NFVO检查资源池中资源情况，能否满足此次实例化的需求；若不满足，则授权失败；否则执行S07；
S07，NFVO向VIM发起资源预留请求。
S08，VIM为此实例化需求预留相应资源，并向NFVO回复响应(携带VIM标识)。
S09，VFVO向VNFM返回实例化VNF操作授权响应。
S10，VNFM转发此响应到EMS。EMS按S03确定的VDU列表，将其转成相应的部署VDU的命令列表。举例如下，EMS生成的部署VDU的命令列表如下：
第1条：ADD VDU：VDU_1(VDU编号)，VDU_A(VDU类型)，…
第2条：ADD VDU:VDU_2(VDU编号)，VDU_A(VDU类型)，…
..
S11，EMS分别执行每一条命令(比如第1条命令)，内部先产生对VM规格和链路的需求，向VNFM发起资源分配请求。
S12，VNFM向S08确定的VIM发起资源分配请求。
S13，VIM为此请求分配VM_A类型VM_1，并根据请求中的链路需求，为VM_1与周边VM建立虚拟网络连接，分配成功返回VM索引(VM_1)。
S14-S15，VNFM向EMS转发此资源分配响应。
S16，EMS在VM_1的基础上，部署VDU_A的软件，完成一些基础数据配置，在EMS及VNF的数据表中记录下VDU_1的基础数据配置信息作为资源配置信息，以便EMS后续可以对VDU_1进行管理。
S17，EM循环执行每一个配置命令(即循环执行S11-S16)完成每个所需VDU的部署，待所有VDU部署完成后，通知VNFM实例化完成。
S18，VNFM告知NFVO实例化完成，NFVO刷新此实例的资源使用等信息。
具体实现中，VNF实例扩容的过程与VNF实例化过程类似，其不同在于，处理的VDU的个数不同以及确定是否扩容的策略不同，具体的扩容策略在前面的实施例已经介绍。因此，在此，不对VNF实例扩容流程进行场景描述。
图39为本发明的终止VNF实例过程的原理流程示意图。如图39所示，其可包括：
S20，VNFM收到终止VNF实例请求，请求方为NFVO；
S21，VNFM将请求转发到相应的EMS。
S22，EMS根据存储的资源配置信息，确定需终止的VNF及其包括的VDU列表，并生成删除VDU的命令列表，所述列表中包括对应于所述VNF实例的需删除的VDU的删除命令。
S23，EMS向VNF发起卸载某个VDU的请求，VNF卸载此VDU对应的软件。
S24，VDU软件卸载完成后，EMS执行VDU的删除命令，将EMS上与此VDU相关的基础数据删除。
S25，同时EMS向VNFM发起资源释放请求，请求释放VDU相关的VM 和链路。
具体实现中，S24和S25不分先后，但要保证资源释放请求中能带上VM的标识，以便VIM能正确删除。
S26，VNFM转发此请求到相应的VIM。
S27-S28，VIM释放此VM和相关联的链路资源，并回复资源释放响应。
S29，VNFM转发此响应到EMS。
EMS循环执行S23～S29，将需终止的VNF内所有VDU及其资源释放完成。如果EMS上还有此VNF相关的数据，也进行删除。
S30，EMS向VNFM回复VNF实例终止完成响应。
S31，VNFM删除此VNF实例相关的数据，并向NFVO回复VNF实例终止完成响应，NFVO也删除此VNF实例相关的数据信息。
具体实现中，VNF实例缩容的过程与终止VNF实例的过程类似，其不同在于，处理的VDU的个数不同以及确定是否缩容的策略不同，具体的缩容策略在前面的实施例已经介绍。因此，在此，不对VNF实例缩容流程进行场景描述。
另外，本发明实施例还提供了一种计算机存储介质，该计算机存储介质可存储有程序，该程序执行时可运行本发明实施例所述的方法的部分或全部步骤。具体实现中，本发明实施例的计算机存储介质包括：RAM、ROM、EEPROM、闪存、CD-ROM、DVD或其他光存储器，磁带、磁盘或其他磁存储器，或者其他任何可以用于存储所需信息并可被计算机设备所访问的介质。
由此可见，本发明实施例其相对于现有技术中VNFM(可为图1所示架构中的VNFM)完全依赖VNFD执行相应的VNF生命周期管理具有如下优势：
不再要求各Vendor提供的VNFD覆盖各个生命周期管理全部过程，也不需要VNFD做到标准化，不同的Vendor的网元管理设备之间能体现各Vendor的差异性。并且，当网元管理设备收到一个生命周期(Lifecycle)管理的操作后，不再需要严格根据VNFD中相应参数或流程、脚本的指示，一步一步完成相应的动作，使得VNF的生命周期管理过程非常灵活性。
经过本发明实施例的方法之后，最终VNFD可简化如下：
简化方式1：VNFD中仅保留vendor(对应表1中参数列的vendor及其对应的参数值)、VNF类型(对应表1中参数列的VNF id及其对应的参数值)、版本(对应表1中参数列的VNF版本及其对应的参数值)、容量规格(对应表1中参数列的容量及其对应的参数值)等基本信息。而现有技术VNFD中的其他信息均不需要保留。
简化方式2：将整个VNFD去除，但需将NSD模板做适当修改，确保原来通过关联VNFD能得到的Vendor、VNF类型、版本信息，容量规格等能够在NSD中体现。
总之，不论采用哪种简化方式，只要执行本发明实施例的方法时NFVO->VNFM的接口包含如下信息即可：
Vendor+VNF类型：用于VNFM判断将请求转发到正确的网元管理设备；
VNF类型、VNF版本号、容量规格：由VNFM转发到网元管理设备，以使网元管理设备用此决策实例化哪个版本的VNF，按多大的规格来部署。
具体实现中，当执行VNF生命周期管理的实体为LMU时，本发明实施例还需VNFD或NSD中需保留所述LMU的参数和脚本，进而使得本发明实施例在通过LMU执行VNF生命周期管理之前，能首先将LMU部署起来。具体实现中，LMU的部署流程与现有技术中通过VNFD部署VDU流程一致，在此不进行赘述。
显然，本领域的技术人员可以对本发明进行各种改动和变型而不脱离本发明的精神和范围。这样，倘且本发明的这些修改和变型属于本发明权利要求及其等同技术的范围之内，则本发明也意图包含这些改动和变型在内。
以上所列举的仅为本发明较佳实施例而已，当然不能以此来限定本发明之权利范围，因此依本发明权利要求所作的等同变化，仍属本发明所涵盖的范围。

Claims (16)

  1. 一种实例化虚拟化的网络功能VNF的方法，其特征在于，包括：
    网元管理设备接收VNF管理器VNFM发送的实例化VNF请求消息；
    所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源；
    所述网元管理设备根据所述确定出的资源来实例化VNF。
  2. 如权利要求1所述的方法，其特征在于，
    所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源具体包括：
    所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定实例化VNF所需的虚拟部署单元VDU列表，以及所述VDU列表中各VDU所需虚拟机VM资源；
    所述网元管理设备根据所述确定出的资源来实例化VNF具体包括：
    所述网元管理设备根据所述确定的VDU列表生成部署VDU的命令列表；
    所述网元管理设备在执行所述命令列表中的VDU的部署命令过程中，向虚拟化基础设施管理系统VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置；
    所述网元管理设备在所述VDU列表中所有VDU的部署命令执行完之后，通知所述VNFM所述实例化VNF完成。
  3. 如权利要求2所述的方法，其特征在于，
    所述网元管理设备在执行所述命令列表中的VDU的部署命令过程中，向所述VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置具体包括：
    所述网元管理设备逐一执行所述命令列表中的每一个VDU的部署命令，在执行每一个VDU的部署命令过程中，向所述VIM请求分配所述每一个VDU所需的虚拟机VM资源，并根据所述VIM为所述每一个VDU分配的资源为所述每一个VDU加载软件，并根据所述VIM对每一个VDU分配所需的虚拟机VM资源的分配结果，为所述每一个VDU加载软件并进行基础数据配置。
  4. 如权利要求2或3所述的方法，其特征在于，
    所述网元管理设备根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定实例化VNF所需的虚拟部署单元VDU列表具体包括：
    所述网元管理设备根据所述实例化VNF的请求确定要部署的VNF的类型、版本号及容量规格；
    所述网元管理设备根据所述资源配置信息中的容量规格与VDU的配比关系信息，确定所述要部署的VNF所需的VDU列表。
  5. 如权利要求4所述的方法，其特征在于，所述网元管理设备根据所述实例化VNF的请求确定要部署的VNF的类型、版本号及容量规格，包括：
    所述网元管理设备根据所述实例化VNF请求中携带的VNF的信息模型描述模版VNFD索引信息和容量规格，到对应的VNFD中查询要部署的VNF的类型、版本号和容量规格；
    或者，所述网元管理设备根据所述实例化VNF请求携带的VNF的类型、版本号及容量规格确定要部署的VNF的类型、版本号及容量规格。
  6. 如权利要求2-5中任一项所述的方法，其特征在于，
    所述网元管理设备根据所述确定的VDU列表，生成部署VDU的命令列表之前，还包括：
    所述网元管理设备经由所述VNFM向网络功能虚拟化编排器NFVO发起实例化VNF的操作授权请求消息，并经所述VNFM从所述NFVO接收所述实例 化VNF的操作授权响应消息，所述实例化VNF的操作授权响应消息是所述NFVO确定存在执行所述实例化VNF所需的资源并请求所述VIM预留相应资源后发送的。
  7. 如权利要求1-6中任一项所述的方法，其特征在于，还包括：
    所述网元管理设备获取VNF上报的性能参数；
    所述网元管理设备根据所述上报的性能参数及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，所述要终止的VDU列表中包括一个或多个需终止的VDU，所述资源配置信息中的VNF实例缩容策略包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止一个或多个指定类型的VDU；
    所述网元管理设备根据所述确定的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例缩容。
  8. 如权利要求1-7中任一项所述的方法，其特征在于，所述网元管理设备包括网元管理系统EMS或VNF本地管理单元LMU。
  9. 一种网元管理设备，其特征在于，包括：
    接收模块，用于接收VNF管理器VNFM发送的实例化VNF请求消息；
    确定模块，用于根据所述接收模块接收的所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定执行实例化VNF所需的资源；
    实例化模块，用于根据所述确定模块所确定出的资源来实例化VNF。
  10. 如权利要求9所述的网元管理设备，其特征在于，
    所述确定模块具体用于根据所述实例化VNF请求消息和所述网元管理设备中存储的资源配置信息确定实例化VNF所需的虚拟部署单元VDU列表，以及 所述VDU列表中各VDU所需虚拟机VM资源；
    所述实例化模块具体包括：
    生成模块，用于根据所述确定模块所确定的VDU列表生成部署VDU的命令列表；
    部署模块，用于执行所述命令列表中的VDU的部署命令，并在执行所述命令列表中的VDU的部署命令过程中，向虚拟化基础设施管理系统VIM请求所述命令列表中各VDU所需的虚拟机VM资源，并根据所述VIM对所述命令列表中各VDU所需的虚拟机VM资源的分配结果，为所述VDU列表中的各VDU加载软件及进行基础数据配置；
    通知模块，用于在所述部署模块将所述VDU列表中所有VDU的部署命令执行完之后，通知所述VNFM所述实例化VNF完成。
  11. 如权利要求10所述的网元管理设备，其特征在于，
    所述部署模块具体用于逐一执行所述命令列表中的每一个VDU的部署命令，在执行每一个VDU的部署命令过程中，向所述VIM请求分配所述每一个VDU所需的虚拟机VM资源，并根据所述VIM为所述每一个VDU分配的资源为所述每一个VDU加载软件，并根据所述VIM对每一个VDU分配所需的虚拟机VM资源的分配结果，为所述每一个VDU加载软件并进行基础数据配置。
  12. 如权利要求10或11所述的网元管理设备，其特征在于，
    所述确定模块，具体用于根据所述接收模块所接收的实例化VNF的请求确定要部署的VNF的类型、版本号及容量规格；以及，根据所述网元管理设备存储的资源配置信息中的容量规格与VDU的配比关系信息，确定所述要部署的VNF所需的VDU列表；以及确定所述VDU列表中各VDU所需虚拟机VM资源。
  13. 如权利要求12所述的网元管理设备，其特征在于，所述根据所述实例 化VNF的请求确定要部署的VNF的类型、版本号及容量规格，具体包括：根据所述接收模块接收的实例化VNF请求中携带的VNF的信息模型描述模版VNFD索引信息和容量规格，到对应的VNFD中查询要部署的VNF的类型、版本号和容量规格；或者，根据所述接收模块接收的实例化VNF请求携带的VNF的类型、版本号及容量规格确定要部署的VNF的类型、版本号及容量规格。
  14. 如权利要求10-13中任一项所述的网元管理设备，其特征在于，还包括：
    操作授权处理模块，用于经由所述VNFM向网络功能虚拟化编排器NFVO发起实例化VNF的操作授权请求消息，并经所述VNFM从所述NFVO接收所述实例化VNF的操作授权响应消息，所述实例化VNF的操作授权响应消息是所述NFVO确定存在执行所述实例化VNF所需的资源并请求所述VIM预留相应资源后发送的。
  15. 如权利要求9-14中任一项所述的网元管理设备，其特征在于，还包括：
    获取模块，用于获取VNF上报的性能参数；
    所述确定模块还用于根据所述上报的性能参数及所述资源配置信息中的VNF实例缩容策略，确定执行VNF实例缩容要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，所述要终止的VDU列表中包括一个或多个需终止的VDU，所述资源配置信息中的VNF实例缩容策略包括：当上一次上报的性能参数大于第一阈值，本次上报的性能参数小于第一阈值与一缓冲参数的差值时，执行VNF实例缩容，终止一个或多个指定类型的VDU；
    所述网元管理设备还包括：
    缩容模块，用于根据所述确定的要终止的VDU列表及所述VDU列表中各VDU对应的需释放的VM资源，执行VNF实例缩容。
  16. 如权利要求9-15中任一项所述的网元管理设备，其特征在于，包括网元管理系统EMS或VNF本地管理单元LMU。
PCT/CN2015/092869 2014-11-06 2015-10-26 一种实例化vnf的方法及网元管理设备 WO2016070729A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410621561.3 2014-11-06
CN201410621561.3A CN105634782B (zh) 2014-11-06 2014-11-06 一种实例化vnf的方法及网元管理设备

Publications (1)

Publication Number Publication Date
WO2016070729A1 true WO2016070729A1 (zh) 2016-05-12

Family

ID=55908560

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/092869 WO2016070729A1 (zh) 2014-11-06 2015-10-26 一种实例化vnf的方法及网元管理设备

Country Status (2)

Country Link
CN (1) CN105634782B (zh)
WO (1) WO2016070729A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108429631A (zh) * 2017-02-15 2018-08-21 华为技术有限公司 一种网络业务实例化的方法及装置
CN110115012A (zh) * 2016-12-30 2019-08-09 华为技术有限公司 一种秘密信息的分发方法和设备

Families Citing this family (23)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105791147B (zh) * 2014-12-26 2019-07-23 华为技术有限公司 负载均衡器的连接方法、业务的实例化部署方法及装置
WO2017067616A1 (en) * 2015-10-23 2017-04-27 Telefonaktiebolaget Lm Ericsson (Publ) Allocating hosts for instances with anti affinity rule
CN107493184B (zh) * 2016-06-12 2022-06-07 中兴通讯股份有限公司 虚拟化网络功能部署方法及系统
EP3476080A4 (en) * 2016-06-23 2019-11-13 Intel IP Corporation DEVICE AND METHOD FOR NFV LIFE CYCLE MANAGEMENT
CN109417488B (zh) * 2016-06-28 2020-10-23 华为技术有限公司 虚拟网络功能资源管理的方法和设备
CN107689882B (zh) * 2016-08-05 2020-04-21 华为技术有限公司 一种虚拟化网络中业务部署的方法和装置
CN107770067B (zh) * 2016-08-23 2021-05-11 中兴通讯股份有限公司 消息发送方法和装置
CN107786360A (zh) * 2016-08-30 2018-03-09 中兴通讯股份有限公司 网络功能虚拟化管理方法及装置
CN107819598A (zh) * 2016-09-13 2018-03-20 中兴通讯股份有限公司 一种管理网络功能节点的方法及装置
CN107959582B (zh) * 2016-10-18 2020-12-01 华为技术有限公司 一种切片实例的管理方法及装置
CN108023749A (zh) * 2016-11-03 2018-05-11 华为技术有限公司 管理网络切片的方法、设备和系统
CN108132827B (zh) * 2016-11-30 2021-12-10 华为技术有限公司 一种网络切片资源映射方法、相关设备及系统
CN109213567B (zh) 2017-06-29 2022-05-13 华为技术有限公司 一种管理vnf实例化的方法和设备
EP3704575A4 (en) * 2017-10-31 2021-08-25 Nokia Solutions and Networks Oy REAL-TIME VIRTUAL NETWORK FUNCTIONAL ORCHESTRATION METHOD, APPARATUS AND SYSTEM
CN108011846B (zh) 2017-11-03 2022-07-26 华为技术有限公司 网络功能虚拟化架构中管理业务的方法及装置
CN110311798B (zh) * 2018-03-20 2021-02-12 华为技术有限公司 一种管理虚拟资源的方法及装置
WO2020077585A1 (zh) * 2018-10-18 2020-04-23 华为技术有限公司 Vnf服务实例化方法及装置
CN109379699B (zh) * 2018-11-12 2020-08-25 中国联合网络通信集团有限公司 创建虚拟化转发面网元的方法及装置
CN109547252A (zh) * 2018-11-27 2019-03-29 新华三技术有限公司合肥分公司 实例化节点的方法、装置和服务器
CN111447076B (zh) * 2019-01-17 2023-01-03 中国移动通信有限公司研究院 网络功能虚拟化nvf系统的容器部署方法及网元
CN111641515B (zh) * 2019-03-01 2021-11-19 华为技术有限公司 Vnf的生命周期管理方法及装置
CN113127138B (zh) * 2019-12-31 2024-04-16 西安华为技术有限公司 虚拟化网络功能部署方法、装置和系统
CN113726540A (zh) * 2020-05-25 2021-11-30 中移(苏州)软件技术有限公司 一种网元处理方法、装置、电子设备和存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2007037867A2 (en) * 2005-09-15 2007-04-05 Tp Lab, Inc. Method to dynamically create a virtual network
CN104009871A (zh) * 2014-06-06 2014-08-27 中国科学院声学研究所 Sdn控制器实现方法及sdn控制器
CN104050045A (zh) * 2014-06-27 2014-09-17 华为技术有限公司 基于磁盘io的虚拟资源分配方法及装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102156665B (zh) * 2011-04-13 2012-12-05 杭州电子科技大学 一种虚拟化系统竞争资源差异化服务方法
US9621425B2 (en) * 2013-03-27 2017-04-11 Telefonaktiebolaget L M Ericsson Method and system to allocate bandwidth for heterogeneous bandwidth request in cloud computing networks
CN103607349B (zh) * 2013-11-14 2017-02-22 华为技术有限公司 虚拟网络中确定路由的方法及运营商边缘设备

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2007037867A2 (en) * 2005-09-15 2007-04-05 Tp Lab, Inc. Method to dynamically create a virtual network
CN104009871A (zh) * 2014-06-06 2014-08-27 中国科学院声学研究所 Sdn控制器实现方法及sdn控制器
CN104050045A (zh) * 2014-06-27 2014-09-17 华为技术有限公司 基于磁盘io的虚拟资源分配方法及装置

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110115012A (zh) * 2016-12-30 2019-08-09 华为技术有限公司 一种秘密信息的分发方法和设备
CN110115012B (zh) * 2016-12-30 2020-12-25 华为技术有限公司 一种秘密信息的分发方法和设备
US11025594B2 (en) 2016-12-30 2021-06-01 Huawei Technologies Co., Ltd. Secret information distribution method and device
CN108429631A (zh) * 2017-02-15 2018-08-21 华为技术有限公司 一种网络业务实例化的方法及装置
WO2018149263A1 (zh) * 2017-02-15 2018-08-23 华为技术有限公司 一种网络业务实例化的方法及装置
CN108429631B (zh) * 2017-02-15 2021-01-15 华为技术有限公司 一种网络业务实例化的方法及装置

Also Published As

Publication number Publication date
CN105634782A (zh) 2016-06-01
CN105634782B (zh) 2019-03-01

Similar Documents

Publication Publication Date Title
WO2016070729A1 (zh) 一种实例化vnf的方法及网元管理设备
US10326653B2 (en) Method for upgrading network functions virtualization application, service forwarding method, and apparatus
WO2019233273A1 (zh) 管理容器服务的方法和装置
WO2020103465A1 (zh) 统一资源调度协调器及其创建虚拟机和/或容器的方法、统一资源调度系统
WO2018024059A1 (zh) 一种虚拟化网络中业务部署的方法和装置
WO2017012381A1 (zh) 一种生命周期管理方法及装置
US20220075666A1 (en) Containerized vnf deployment method and related device
WO2020135799A1 (zh) Vnf服务实例化方法及装置
WO2016165292A1 (zh) 一种实现虚拟网络功能部署规格配置的方法及装置
US20230006944A1 (en) Interoperable cloud based media processing using dynamic network interface
WO2016131172A1 (zh) 一种更新网络服务描述器nsd的方法及装置
WO2020103925A1 (zh) 一种容器化虚拟网络功能的部署方法和装置
US20180121227A1 (en) Hardware acceleration method and related device
WO2020011214A1 (zh) 管理虚拟化资源的方法和装置
WO2019179301A1 (zh) 一种管理虚拟资源的方法及装置
JP7233560B2 (ja) ワークフローを処理する方法及び装置
WO2019129118A1 (zh) Vnf服务实例化方法及装置
WO2016062116A1 (zh) 一种虚拟化网络备份、恢复的方法和相应装置
CN114467076A (zh) 用于步进工作流的方法和装置
WO2019047835A1 (zh) 虚拟网络功能的实例化方法
WO2020015751A1 (zh) 容器服务快照的管理方法和装置
WO2017011938A1 (zh) 虚拟网络功能扩容的方法和装置
WO2022140945A1 (zh) 容器集群的管理方法和装置
JP6801776B2 (ja) 仮想ネットワーク機能管理装置、仮想インフラストラクチャ管理装置、及び仮想ネットワーク機能構築方法
WO2016115879A1 (zh) 虚拟化网络功能与网元管理系统建立连接的方法和装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 15856183

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15856183

Country of ref document: EP

Kind code of ref document: A1