CN108345490B - Method and system for deploying virtual machine in NFV - Google Patents

Method and system for deploying virtual machine in NFV Download PDF

Info

Publication number
CN108345490B
CN108345490B CN201710054079.XA CN201710054079A CN108345490B CN 108345490 B CN108345490 B CN 108345490B CN 201710054079 A CN201710054079 A CN 201710054079A CN 108345490 B CN108345490 B CN 108345490B
Authority
CN
China
Prior art keywords
node
virtual machine
resource
appeal
bandwidth
Prior art date
Legal status (The legal status 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 status listed.)
Active
Application number
CN201710054079.XA
Other languages
Chinese (zh)
Other versions
CN108345490A (en
Inventor
杨永法
田彦峰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
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 Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to CN201710054079.XA priority Critical patent/CN108345490B/en
Publication of CN108345490A publication Critical patent/CN108345490A/en
Application granted granted Critical
Publication of CN108345490B publication Critical patent/CN108345490B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5061Partitioning or combining of resources
    • G06F9/5077Logical partitioning of resources; Management or configuration of virtualized resources
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance or administration or management of packet switching networks
    • H04L41/08Configuration management of network or network elements
    • H04L41/0896Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities, e.g. bandwidth on demand
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/455Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
    • G06F9/45533Hypervisors; Virtual machine monitors
    • G06F9/45558Hypervisor-specific management and integration aspects
    • G06F2009/45562Creating, deleting, cloning virtual machine instances

Abstract

The application discloses a method and a system for controlling deployment positions of virtual machines in NFV, which belong to the field of communication and are used for realizing the ordered allocation of bandwidth resources among the virtual machines and avoiding mutual occupation of the bandwidth resources among the virtual machines. The method comprises the following steps: the VNFM acquires resource appeal of each VM through the VNFD, and transmits the resource appeal of each VM to the VIM, wherein the resource appeal comprises bandwidth resource appeal; and the VIM determines the deployment position of each VM according to the resource appeal of each VM, so that the sum of the bandwidth resource appeal of the VMs deployed on each node is smaller than the upper limit of the node bandwidth. The method and the device are used for virtual machine deployment.

Description

Method and system for deploying virtual machine in NFV
Technical Field
The present application relates to the field of communications, and in particular, to a method and a system for deploying a virtual machine in an NFV.
Background
Network Function Virtualization (NFV) technology can be simply understood as migrating the functions of various Network elements used in a telecommunication Network from the current dedicated hardware platform to a common Commercial-off-the-shelf (COTS) server. By means of the NFV technology, each network element used in the telecommunication network can be converted into independent application and flexibly deployed on a unified infrastructure platform constructed by devices such as a server, a storage and a switch. Meanwhile, by means of a virtualization technology, resources of infrastructure hardware equipment are pooled and virtualized, virtual resources are provided for upper-layer applications, application and hardware decoupling is achieved, each application can rapidly increase the virtual resources to achieve the purpose of rapidly expanding the system capacity, or each application can rapidly reduce the virtual resources to achieve the purpose of shrinking the system capacity, and network elasticity is greatly improved.
In the related art, when a Virtual Machine (VM) is deployed in an NFV environment, a situation may occur in which Virtual machines with different bandwidth demands of different Virtual Network Functions (VNFs) are deployed on the same node (blade)/host (host), so that a sum of bandwidth resource demands of the VMs on a single node may exceed a bandwidth upper limit of the node, and bandwidth resources may be mutually preempted among the Virtual machines.
Disclosure of Invention
The embodiment of the invention provides a method and a system for deploying virtual machines in NFV (network virtual machine), which are used for realizing the ordered allocation of the bandwidth resource occupation among the virtual machines and avoiding the mutual occupation of the bandwidth resources among the virtual machines.
In a first aspect, the present application provides a method for deploying a virtual machine in an NFV. As a possible implementation manner of the present application, the method includes:
a Virtual Network Function Manager (VNF Manager, VNFM) acquires resource appeal information of each Virtual machine through a Virtual Network Function Descriptor (VNFD), and transmits the resource appeal information of each VM to a Virtualized Infrastructure Manager (VIM), wherein the resource appeal information includes bandwidth resource appeal information;
and the VIM deploys each VM according to the resource appeal information of each VM, so that the sum of the bandwidth resource appeal of the VMs deployed on each node is smaller than the bandwidth upper limit.
In the application, the VNFM acquires the bandwidth resource appeal of each VM through the VNFD and transmits the bandwidth resource appeal to the VIM, so that the VIM can consider the bandwidth resource appeal of each VM when deploying the VMs, the sum of the bandwidth resource appeal of the VMs deployed on each node is ensured to be smaller than the bandwidth upper limit of the node, the ordered allocation of the bandwidth resources among the virtual machines is realized, and the mutual occupation of the bandwidth resources among the virtual machines is avoided.
With reference to the first aspect, in a first possible implementation manner, the deploying, by the VIM, each VM according to the resource appeal information of each VM may include:
the VIM deploys each VM on each node in a bandwidth balancing mode according to the bandwidth resource appeal information of each VM, so that the difference value of the sum of the bandwidth resource appeal of the VMs deployed on each node is smaller than a threshold value;
or the VIM deploys each VM on each node in a node compact deployment manner according to the bandwidth resource appeal information of each VM, so that the VM is continuously deployed on the next node of the node only when the bandwidth resource on one node cannot continuously accommodate the VM.
In the embodiment of the invention, the bandwidth balance deployment mode is adopted to ensure that the bandwidth occupation of the virtual machines on each node is relatively balanced, and the operation stability of each node in the cluster is improved. And a compact deployment mode is adopted, so that the maximum reasonable utilization of bandwidth resources on the nodes can be realized.
The method for deploying the virtual machine in the NFV can be applied to various scenes (such as VM elasticity increase, VM migration, VM fault reconstruction and the like) in the VM life cycle, and in the scenes, the bandwidth resource is always used as the basic attribute (treated as the same as CPU resource and memory resource) of the VM to determine the deployment position of the VM. Therefore, the problem that when multiple VNFs are deployed in a common cluster, for example, the VMs deployed on one node/host have no bandwidth control due to no bandwidth control, so that bandwidth resources are mutually occupied among the VMs, and finally the VMs cannot reach the expected service specification is solved.
With reference to the first aspect, in a second possible implementation manner, the method further includes:
when the virtual machines are increased, the VIM deploys the increased virtual machines on the nodes meeting the bandwidth resource requirements of the increased virtual machines according to the bandwidth resource occupation condition of each node.
The embodiment of the invention can realize the orderly distribution of the occupation of the bandwidth resources among the virtual machines under the scene of elastically increasing and deleting the virtual machines, and avoid the mutual occupation of the bandwidth resources among the virtual machines.
With reference to the first aspect, in a third possible implementation manner, the method further includes:
when virtual machine migration occurs, the VIM automatically deploys the virtual machines to be migrated on the nodes meeting the bandwidth resource requirements of the virtual machines to be migrated according to the bandwidth resource occupation condition of each node;
alternatively, the first and second electrodes may be,
when the virtual machine is migrated, the VIM judges whether a node designated by a user meets the bandwidth resource demand of the virtual machine needing migration; if yes, migrating the virtual machine needing to be migrated to the node specified by the user; if not, the migration is not performed.
The embodiment of the invention can realize the orderly distribution of the occupation of the bandwidth resources among the virtual machines under the migration scene of the virtual machines, and avoid the mutual occupation of the bandwidth resources among the virtual machines.
With reference to the first aspect, in a fourth possible implementation manner, the method further includes:
when a node with a virtual machine is in fault, for each virtual machine on the node in fault, the VIM reconstructs the virtual machine on the node meeting the bandwidth resource demand of the virtual machine according to the bandwidth resource occupation condition of each node.
The embodiment of the invention can realize the orderly distribution of the occupation of the bandwidth resources among the virtual machines under the reconstruction scene of the virtual machines, and avoid the mutual occupation of the bandwidth resources among the virtual machines.
In a second aspect, the present application provides a system for deploying virtual machines in NFV, the system comprising VNFM and VIM, wherein: the VNFM may be the VNFM mentioned in the first aspect or any one of the possible implementations of the first aspect, and is configured to perform corresponding steps in an embodiment of a method for deploying a virtual machine in an NFV; the VIM may be the VIM mentioned in the first aspect or any one of the possible implementation manners of the first aspect, and is configured to perform corresponding steps in an embodiment of a method for deploying a virtual machine in an NFV.
In the first and second aspects of the present application, the resource appeal further includes at least one of: CPU resource appeal, memory resource appeal and hard disk resource appeal.
In a third aspect, the present application provides a computer program product which, when run on a computer, causes the computer to perform the method for deploying a virtual machine in an NFV as provided in any of the embodiments of the present application above.
In a fourth aspect, the present application provides a computer-readable storage medium (including both temporary and non-temporary storage media) having a computer program stored thereon, which, when executed by a processor, performs a method for deploying a virtual machine in an NFV as provided in any of the embodiments of the present application.
Drawings
Fig. 1 is a system architecture diagram of an NFV provided by an embodiment of the present invention;
fig. 2 is a flowchart of a method for deploying a virtual machine in an NFV according to an embodiment of the present invention;
fig. 3 is a schematic diagram of controlling deployment locations of VMs according to bandwidth appeal of VMs under an initial deployment scenario;
fig. 4 is a schematic diagram of controlling deployment locations of VMs according to bandwidth appeal of VMs under a resilient scenario;
fig. 5 is a schematic diagram of controlling deployment locations of VMs according to bandwidth appeal of the VMs in a virtual machine migration scenario;
fig. 6 is a schematic diagram illustrating a deployment location of a VM controlled according to a bandwidth appeal of the VM in a remote reconstruction scenario of the virtual machine;
fig. 7 is a schematic diagram of a system for deploying virtual machines in an NFV according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
Fig. 1 is a system architecture diagram of NFV. The NFV system 100 may be implemented in various networks, such as in a data center network, an operator network, or a local area network. Referring to fig. 1, the NFV System 100 includes an NFV Management and Orchestration System (NFVMANO)101, an NFV Infrastructure layer (NFVI) 130, a plurality of Virtual Network Functions (VNFs) 108, a plurality of Element Management (EM) 122, a Network Service, a VNF and Infrastructure Description (Network Service, VNF and Infrastructure Description)126, and an Operation-Support Management System (OSS/BSS) 124.
Among other things, the NFV management and orchestration system 101 includes an NFV Orchestrator (NFV editor, NFVO)102, one or more virtual network function managers (VNF managers, VNFM)104, and a Virtualized Infrastructure Manager (VIM) 106.
NFV management and orchestration system (NFVMANO)101 is used to perform monitoring and management of VNF108 and NFVI 130. The NFVO 102 may implement network services on the NFVI 130, and may also perform resource-related requests from one or more VNFMs 104, send configuration information to the VNFMs 104, and collect status information of the VNFs 108. Additionally, NFVO 102 may communicate with VIM106 to enable allocation and/or reservation of resources and exchange configuration and status information for virtualized hardware resources. The VNFM104 may manage one or more VNFs 108. The VNFM104 may perform various management functions such as instantiating, updating, querying, scaling, and/or terminating the VNF108, among others. VIM106 may perform functions for resource management, such as managing allocation of infrastructure resources (e.g., adding resources to virtual containers) and operational functions (e.g., collecting NFVI fault information). The VNFM104 and VIM106 may communicate with each other for resource allocation and exchange configuration and status information for virtualized hardware resources.
A Virtual Network Function (VNF) may be deployed in an implementation of a Network Function module of the NFVI.
NFVI 130 includes computing hardware 112, storage hardware 114, network hardware 116, Virtualization Layer (Virtualization Layer), virtual computing 110, virtual storage 118, and virtual network 120. The network services, VNF and infrastructure description 126 and OSS/BSS 124 are further discussed in the European Telecommunications Standards Institute (ETSI) Group Specification (GS) NFV 002V1.1.1 standard.
The NFVI 130 includes hardware resources, software resources, or a combination of both to complete the deployment of the virtualized environment. In other words, the hardware resources and virtualization layer are used to provide virtualized resources, e.g., as virtual machines and other forms of virtual containers, for VNF 108. The hardware resources include computing hardware 112, storage hardware 114, and network hardware 116. The computing hardware 112 may be commercially available hardware and/or custom hardware to provide processing and computing resources. The storage hardware 114 may be storage capacity provided within a network or storage capacity residing within the storage hardware 114 itself (local storage located within a server). In one implementation, the resources of the computing hardware 112 and the storage hardware 114 may be pooled together. The network hardware 116 may be a switch, a router, and/or any other network device configured with switching functionality. The network hardware 116 may span multiple domains and may include multiple networks interconnected by one or more transport networks.
A virtualization layer within NFVI 130 may abstract hardware resources from the physical layer and decouple VNF108 to provide virtualized resources to VNF 108. The virtual resource layer includes virtual compute 110, virtual memory 118, and virtual network 120. Virtual compute 110 and virtual storage 118 may be provided to VNF108 in the form of virtual machines, and/or other virtual containers. The virtualization layer abstracts the network hardware 116 to form a Virtual network 120, and the Virtual network 120 may include Virtual switches (Virtual switches) that are used to provide connections between Virtual machines and other Virtual machines. In addition, the transport network in the network hardware 116 may be virtualized using a centralized control plane and a separate forwarding plane (e.g., a software defined network).
As shown in fig. 1, VNFM104 may interact with VNF108 and EM 122 to manage the VNF's lifecycle and exchange configuration and status information. VNF108 may be configured to perform virtualization of at least one network function through one physical network device. For example, in one implementation, the VNF108 may be configured to provide functions provided by different network elements in an IP Multimedia Subsystem (IMS) network, such as network functions of a Serving Call Session Control Function (S-CSCF) or a Home Subscriber Server (HSS), and so on. EM 122 is configured to manage one or more VNFs 108.
Fig. 2 is a flowchart of a method for deploying a virtual machine in an NFV according to an embodiment of the present invention. Referring to fig. 2, a method for deploying a virtual machine in an NFV provided in an embodiment of the present invention may include:
21. the VNFM acquires resource appeal of each VM through the VNFD, and transmits resource appeal information of each VM to the VIM, wherein the resource appeal information comprises bandwidth resource appeal information.
Virtual Network Function Descriptor (VNFD): a configuration template describing deployment and operational behavior of a VNF. VNFD is used for the running process of virtualized network function modules and lifecycle management for virtualized network function module instances. One VNF may correspond to one VNFD.
In the embodiment of the present invention, the resource appeal information of each VM may be defined in the VNFD, and the resource appeal information of each VM may include bandwidth resource appeal information, and may further include at least one of CPU resource appeal information, memory resource appeal information, and hard disk resource appeal information, for example. In one case, for example, the VNFD may define bandwidth resource appeal information, CPU resource appeal information, memory resource appeal information, and hard disk resource appeal information for each VM; in another case, for example, the VNFD may define bandwidth resource appeal information, CPU resource appeal information, and memory resource appeal information for each VM. In the embodiment of the present invention, the VNFD defines bandwidth resource appeal information of each VM, and may further define at least one of CPU resource appeal information, memory resource appeal information, and hard disk resource appeal information as needed. The bandwidth resource appeal information of the VM mainly refers to a data volume transmitted in unit time of the VM, for example, megabits per second (Mbps) is taken as a unit, the CPU resource appeal information of the VM mainly refers to the number of cores of the CPU (for example, four-core CPU) required by the VM, the memory resource appeal information of the VM mainly refers to the size of the memory required by the VM, and the hard disk resource appeal information of the VM mainly refers to the size of the hard disk capacity required by the VM.
In the embodiment of the present invention, the VNFD may be, for example, manually imported into the VNFM, so that the VNFM can quickly acquire the resource demand information of each VM. For example, when the VNFM acquires the resource appeal information of the VM corresponding to each VNF, the VNFDs corresponding to each VNF may be acquired in a batch import manner, for example, and then the resource appeal information of each VM is acquired from each VNFD.
After obtaining the resource appeal information of each VM, the VNFM can transmit the obtained resource appeal information of each VM to the VIM.
22. And the VIM deploys each VM according to the resource appeal information of each VM, so that the sum of the bandwidth resource appeal of the VMs deployed on each node is smaller than the bandwidth upper limit.
Wherein the bandwidth upper limit of the node can be determined by the hardware (e.g., network card) capability of the node. If a node for deploying the VM is selected, the bandwidth upper limit corresponding to the node is also determined.
In the embodiment of the invention, the VM can be deployed by adopting a balanced dispersion strategy and a compact deployment strategy in the process of determining the deployment position of the VM according to the resource appeal of each VM.
The method for deploying the VM by using the balanced dispersion strategy may specifically be: the VIM deploys each VM on each node in a bandwidth balancing mode according to the bandwidth resource appeal information of each VM, so that the difference value of the sum of the bandwidth resource appeal of the VMs deployed on each node is smaller than a threshold value, wherein the threshold value can be set according to needs. That is to say, the VM deployment mode using the balanced dispersion strategy ensures that the sum of the bandwidth resource requirements of the VMs deployed on each node does not differ too much, thereby implementing the bandwidth resource on each node. For example, if there are 5 virtual machines (e.g., VM1-VM5, where the bandwidth required by VM1-VM3 is large, the bandwidth required by VM4 and VM5 is small, and the bandwidth required by VM1-VM5 is decreased) and three nodes, when performing virtual machine deployment, one virtual machine may be deployed on each node for virtual machines VM1-VM3, for example, VM1 is deployed on a first node, VM2 is deployed on a second node, and VM3 is deployed on a third node. And for the remaining virtual machines VM4 and VM5 with smaller required bandwidth, deployment is carried out according to the bandwidth resource occupation conditions of the three nodes. For example, VM4 is deployed on the second node and VM5 is deployed on the third node. The final result of the deployment is that the sum of the bandwidth resource appeal of the virtual machines deployed on each node is not too large, and the balance of the bandwidth resources among the nodes is realized.
The manner of deploying the VM by using the compact deployment policy may specifically be: the VIM deploys each VM on each node in a node compact deployment mode according to the bandwidth resource appeal of each VM, so that the VM continues to be deployed on the next node of one node only when the bandwidth resource on the node cannot continuously accommodate the VM. That is to say, when the VM is deployed, the VM is deployed node by node, and only when the current node cannot accommodate any one of the virtual machines that need to be deployed, the VM is continuously deployed at the next node, so that deployment is performed until all the virtual machines are deployed. By adopting the deployment mode, the maximum reasonable utilization of the bandwidth resources on the nodes can be realized. For example, if there are 5 virtual machines (e.g., VM1-VM5, where the bandwidth required by VM1-VM3 is large, the bandwidth required by VM4 and VM5 is small, and the bandwidth required by VM1-VM5 is decreased) and three nodes, when deploying the virtual machines, if the bandwidth upper limit of the first node is greater than the bandwidth requirement of VM1 and VM2, the VM1 and VM2 are deployed on the first node; meanwhile, if the bandwidth upper limit of the second node is larger than the bandwidth appeal of the VM3-VM5, deploying the VM3-VM5 on the second node, and not deploying any virtual machine on the third node; of course, if the bandwidth cap of the second node is not greater than the bandwidth appeal of VM3-VM5 but is only greater than the bandwidth appeal of VM3-VM4, then VM3-VM4 is deployed on the second node and VM5 is deployed on the third node. The final result of the deployment is that the bandwidth resources on each node are fully utilized, and the maximum reasonable utilization of the bandwidth resources is realized.
In the method for controlling deployment positions of virtual machines in NFV provided in the embodiments of the present invention, the VNFM acquires bandwidth resource demand information of each VM through the VNFD and transmits the bandwidth resource demand information to the VIM, so that when the VIM deploys VMs, the bandwidth resource demand information of each VM can be considered, it is ensured that the sum of the bandwidth resource demand information of the VMs deployed on each node is smaller than the node bandwidth upper limit, the ordered allocation of bandwidth resources among virtual machines is realized, and the problem that the VMs cannot reach the service expected specification due to mutual preemption of the bandwidth resources among the virtual machines is avoided.
In the embodiment of the present invention, referring to fig. 1, in an initial deployment stage of the VNF108, the VNF108 may provide VM bandwidth resource appeal information, and when the VNFM104 instantiates the VNF108, the VM bandwidth resource appeal information is also transmitted to the VIM106, so that the VIM106 may treat the VM bandwidth resource appeal as the CPU resource appeal and the memory resource appeal, and comprehensively evaluate a reasonable deployment location of the VM.
In the embodiment of the invention, under the scenes of flexibly adding and deleting a VM scene, a VM migration scene, a node (Blade) fault remote reconstruction VM and the like, the bandwidth resource appeal is always treated as the basic attribute of the VM in the whole life cycle of the VM. Ensuring that the sum of the bandwidth resources of the VM finally deployed on each node does not exceed the bandwidth upper limit of the node. The following further explains these several scenarios with five VNFs as an example, respectively, in conjunction with the drawings.
Initial deployment phase
Referring first to fig. 3, fig. 3 is a schematic diagram illustrating controlling a deployment location of a VM according to a bandwidth appeal of the VM in an initial deployment scenario. In the initial deployment phase, VMs of VNF1, VNF2, VNF3, VNF4, and VNF5 need to share one Cluster (Cluster) deployment. That is, VMs of different VNFs may share a node deployment. When the VMs are deployed, the sum of the bandwidth resources of the VMs on each node is controlled not to exceed the bandwidth upper limit of the node.
Specifically, in the initial state, a specific process of controlling the deployment position of the VM according to the bandwidth appeal of the VM may be as follows:
1) the bandwidth resource appeal of the VM is defined in the VNFD of each VNF. That is, bandwidth resources are defined for each virtual Network Interface Card (vNIC) of each VM: required bandwidth: XXMbps (required _ bandwidth: XXMbps).
2) The VNFM acquires bandwidth resource appeal information of each VM through the VNFD, and when a virtual machine is instantiated, transmits the bandwidth resource appeal information and other resource appeal information such as CPU resource appeal information and memory resource appeal information to the VIM together as basic attributes of the VM.
3) The VIM takes the bandwidth resource appeal information, the CPU resource appeal information and the memory resource appeal information as the basic attributes of the VM, and comprehensively evaluates the reasonable deployment position of the VM.
When the VIM determines a reasonable deployment position of the VM, the control strategies that can be adopted include a balanced decentralized strategy, a compact deployment strategy, and the like. The balanced dispersion strategy may be: the occupation of the bandwidth resources of the VMs deployed on each node is balanced as much as possible, and the purpose of the balanced and dispersed deployment of the large-bandwidth VMs on different hosts may be achieved. The compact deployment strategy may be: the bandwidth resource of a single node is occupied preferentially, but the bandwidth upper limit of the node is not exceeded.
As can be seen from fig. 3, in the initial deployment phase, four virtual machines are deployed on the node 01: VM11, VM31, VM32, and VM 41; two virtual machines are deployed on node 02: VM01 and VM 21; two virtual machines are deployed on the node 03: VM22 and VM 02; node N has deployed thereon a virtual machine VM 12. It should be understood that the particular virtual machine deployment illustrated in FIG. 3 is by way of example only and is not intended to be limiting.
Elastic add-delete VM scene
Referring to fig. 4, fig. 4 is a schematic diagram of controlling deployment positions of VMs according to bandwidth appeal of the VMs in an elastic scenario. Under the condition of the existing VNF 1-VNF 5 co-cluster deployment, if VNF5 needs to flexibly add a virtual machine VM42 due to business needs. In this flexible addition of virtual machines, the selection of VM42 deployment locations needs to consider whether bandwidth resources are sufficient for each node.
Specifically, in the context of elastically increasing a virtual machine, the specific process of controlling the deployment position of the VM according to the bandwidth appeal of the VM may be as follows:
1) the VNFD of the VNF5 defines the bandwidth resource appeal of the virtual machine VM42, and the VNFM acquires the bandwidth resource appeal information of the virtual machine VM42 newly added to the VNF5 through the VNFD.
2) When the VNFM instantiates the virtual machine VM42, the bandwidth resource appeal information, together with other resource appeal information such as CPU resource appeal information, memory resource appeal information, and the like, is delivered to the VIM as a basic attribute of the VM.
3) The VIM comprehensively evaluates reasonable deployment positions of the VMs 42 by using the bandwidth resource appeal information and, for example, CPU resource appeal information, memory resource appeal information, and the like as the VM basic attributes.
4) The VIM determines that the remaining bandwidth resources of node 01 do not meet the bandwidth resource appeal of VM42, and the bandwidth resources of node 02 can meet the bandwidth resource appeal of VM42, so the VIM deploys VM42 on node 02.
As can be seen from fig. 4, fig. 4 adds a new virtual machine VM42 on node 02 based on the deployment location of the cluster virtual machines shown in fig. 3.
Similarly, when the virtual machine VM42 needs to be deleted, the virtual machine VM42 can be deleted directly because the deletion of the virtual machine does not need to consider the bandwidth resources of the node.
It should be understood that the above is only an example of adding one virtual machine, and when a plurality of virtual machines need to be added, the above-described manner may be adopted for each of the added virtual machines.
In the embodiment of the invention, when the virtual machines are increased, the VIM can deploy each increased virtual machine on the node meeting the bandwidth resource appeal of the virtual machine according to the bandwidth resource occupation condition of each node. Therefore, under the scene of elastically increasing and deleting the virtual machines, the orderly distribution of the occupation of the bandwidth resources among the virtual machines can be realized, and the bandwidth resources are prevented from being mutually occupied among the virtual machines.
VM migration scenario
The virtual machine migration in the embodiment of the present invention refers to migrating a virtual machine from one node (for example, one physical server) to another node (for example, another physical server), and ensuring that a service provided by the virtual machine is not interrupted as much as possible. Referring to fig. 5, fig. 5 is a schematic diagram illustrating a deployment location of a VM controlled according to a bandwidth appeal of the VM in a virtual machine migration scenario. Under the condition of the existing VNF 1-VNF 5 co-cluster deployment, if the VM42 newly added on the node 02 in fig. 4 triggers an automatic migration condition, the VIM may consider whether the bandwidth resource of the target node is enough when automatically selecting the target node to which the VM42 migrates. When the VM42 is migrated manually, the VIM needs to determine whether the bandwidth resource of the target node meets the bandwidth requirement of the VM 42.
Specifically, in a virtual machine migration scenario, a specific process of controlling the deployment position of the VM according to the bandwidth appeal of the VM may be as follows:
1) the VM42 triggers a virtual machine migration action, when the VIM selects a target node, the bandwidth resource appeal and the CPU resource appeal, the memory resource appeal and the like are taken as the VM basic attributes, and the node which can meet the resource appeal of the VM42 is comprehensively evaluated.
2) In the case of auto-migration, the VIM automatically selects for migration the nodes that can meet the VM42 bandwidth resource appeal. In this process, if the VIM discovers that the node 01 cannot meet the bandwidth resource appeal of the VM42, the node 03 continues to search for the resource, and the node 03 can meet the bandwidth resource appeal of the VM42 and migrate the VM42 to the node 03.
In the case of manual migration, the VIM determines whether the user-specified node can meet the bandwidth resource appeal of the VM 42. If the node designated by the user cannot meet the bandwidth resource appeal of the VM42, migration fails, and specific reasons are given; if the user-designated node meets the bandwidth resource appeal of the VM42, the VM42 is migrated to the user-designated node.
It should be understood that the above is only an example of migrating one virtual machine, and when a plurality of virtual machines need to be migrated, the above-described manner may be adopted for each of the plurality of virtual machines that need to be migrated.
In the embodiment of the invention, when the virtual machine is migrated, the VIM automatically deploys each virtual machine to be migrated on a node meeting the bandwidth resource appeal of the virtual machine according to the bandwidth resource occupation condition of each node; or when the virtual machine is migrated, the VIM judges whether the node specified by the user meets the bandwidth resource demand of the virtual machine needing to be migrated; if yes, migrating the virtual machine to the node designated by the user; if not, migration fails. Therefore, under the virtual machine migration scene, the sequential allocation of the occupied bandwidth resources among the virtual machines can be realized, and the bandwidth resources are prevented from being mutually occupied among the virtual machines.
Node failure remote reconstruction VM scene
The virtual machine rebuilding in the embodiment of the present invention means that when a virtual machine on one node cannot continue to provide a service (for example, the node fails), the virtual machine on the node is rebuilt on another node to continue to provide the service. Referring to fig. 6, fig. 6 is a schematic diagram illustrating that the deployment position of the VM is controlled according to the bandwidth appeal of the VM in a remote reconstruction scenario of the virtual machine. Under the condition that the existing VNFs 1-5 are deployed in a common cluster and the virtual machines are in the deployment positions shown in fig. 4, if the node 02 where the virtual machine VM42 newly added in fig. 4 is located fails. The virtual machine on node 02 may be rebuilt on other nodes of the same cluster. In the process of selecting the target node, the VIM considers whether the bandwidth resource of the target node is enough.
Specifically, in a scenario of reconstructing the virtual machine in different places, a specific process of controlling the deployment position of the VM according to the bandwidth appeal of the VM may be as follows:
1) and when the VIM detects the failure of the node 02, the virtual machine reconstruction action is triggered. The virtual machines VM01, VM21, VM42 deployed on node 02 need to be rebuilt on off-site nodes.
2) When the VIM selects the target node, the bandwidth resource demand information, the CPU resource demand information, the memory resource demand information, and the like are used as the VM basic attributes, and it is comprehensively evaluated which node can satisfy the resource demand of any one of the VM01, VM21, and VM 42.
3) And the VIM carries out virtual machine reconstruction based on the evaluation result.
In the specific process of virtual machine evaluation and reconstruction, for example, if the node 01 is evaluated by the VIM, and the node 01 is found to be unable to satisfy the resource appeal of any one of VM01, VM21, and VM42, the node 03 is evaluated; if the node 03 can meet the resource requirements (including bandwidth resource requirements, CPU resource requirements and memory resource requirements) of the virtual machine VM42 but cannot meet the resource requirements of the virtual machine VM01 and VM21, reconstructing the virtual machine VM42 on the node 03, and then continuously searching for a node suitable for the virtual machine VM01 and the virtual machine VM 21; when the VIM evaluates the node N, if the node N is found to satisfy the resource requirements of the virtual machines VM01 and VM21, the VM01 and VM21 are rebuilt on the node N.
It should be understood that the above specific node selection manner is only an example to discuss how to select nodes in the context of rebuilding a virtual machine, and is for the purpose of making those skilled in the art better understand the selection process of deployment positions of virtual machines, and is not intended to be limiting.
Meanwhile, it should be understood that the above is only described by taking virtual machine reconstruction for a node that fails as an example, and when a plurality of nodes fail, each of the plurality of nodes that fail may be operated in the above-described manner.
In the embodiment of the invention, when a node with a virtual machine is in fault, the VIM reconstructs each virtual machine on the node in fault on the node meeting the bandwidth resource appeal of the virtual machine according to the bandwidth resource occupation condition of each node. Therefore, under the virtual machine reconstruction scene, the sequential allocation of the occupied bandwidth resources among the virtual machines can be realized, and the bandwidth resources are prevented from being mutually occupied among the virtual machines.
It should also be appreciated that in the various scenarios described above, the VIM may take into account the bandwidth resource appeal and a number of other resource appeal besides the bandwidth resource appeal when determining the deployment location of the virtual machine. In fact, when determining the deployment position of the virtual machine, other resource appeal besides the bandwidth resource appeal can be selected according to needs, and only one of the other resource appeal can be considered, even the other resource appeal is not considered.
To sum up, the embodiment of the present invention provides a method for controlling VM deployment positions based on VM bandwidth resource appeal, where in various scenarios (creation, deletion, migration, fault reconstruction, elasticity, and the like) in a VM life cycle, bandwidth resources are always used as basic attributes (treated as CPU resources and memory resources) of a VM to determine VM deployment positions. The problem that when multiple VNFs are deployed in a common cluster, for example, no bandwidth control guidance causes no bandwidth control on VMs deployed on one node/host, so that bandwidth resources are mutually preempted among the VMs, and finally the VMs cannot reach the expected service specification is solved.
Correspondingly, the embodiment of the invention also provides a system for deploying the virtual machine in the NFV. Fig. 7 is a schematic diagram of a system for deploying virtual machines in an NFV according to an embodiment of the present invention. Referring to fig. 7, a system 700 for deploying virtual machines in NFV provided by an embodiment of the present invention may include a VNFM 701 and a VIM 702. Wherein: the VNFM 701 may be a VNFM mentioned in any embodiment herein, and is configured to perform corresponding steps in an embodiment of a method for deploying a virtual machine in an NFV; the VIM 702 may be the VIM mentioned in any of the embodiments herein, for performing the corresponding steps in the method embodiments for deploying virtual machines in NFV.
The embodiment of the invention provides a system for controlling the deployment position of a VM (virtual machine) based on the bandwidth resource appeal of the VM, wherein in various scenes (such as creation, deletion, migration, fault reconstruction, elasticity and the like) in the life cycle of the VM, the bandwidth resource is always used as the basic attribute (treated as the same as CPU (Central processing Unit) resource and memory resource) of the VM to determine the deployment position of the VM. The problem that when multiple VNFs are deployed in a common cluster, for example, no bandwidth control guidance causes no bandwidth control on VMs deployed on one node/host, so that bandwidth resources are mutually preempted among the VMs, and finally the VMs cannot reach the expected service specification is solved.
The present application may also provide a computer program product which, when run on a computer, causes the computer to perform the method for deploying a virtual machine in an NFV as provided in any of the embodiments of the present application above.
Furthermore, the present application provides a computer readable storage medium (including both temporary and non-temporary storage media) having stored thereon a computer program which, when executed by a processor, executes a method for deploying a virtual machine in an NFV as provided in any of the embodiments of the present application.
It should be noted that: the system for controlling the deployment position of the virtual machine in the NFV and the method embodiment for controlling the deployment position of the virtual machine in the NFV provided by the above embodiments belong to the same concept, and specific implementation processes thereof are detailed in the method embodiment and are not described herein again.
Unless defined otherwise, technical or scientific terms used herein shall have the ordinary meaning as understood by one of ordinary skill in the art to which this application belongs. The use of "first," "second," and similar terms in the description and claims of this application do not denote any order, quantity, or importance, but rather the terms are used to distinguish one element from another. Also, the use of the terms "a" or "an" and the like do not denote a limitation of quantity, but rather denote the presence of at least one. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect.
It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.
The above description is only exemplary embodiments of the present application and should not be taken as limiting the present application, and any modifications, equivalents, improvements and the like that are made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims (13)

1. A method for deploying a virtual machine in a Network Function Virtualization (NFV), the method comprising:
the method comprises the steps that a Virtual Network Function Manager (VNFM) obtains resource appeal information of each virtual machine through a Virtual Network Function Descriptor (VNFD), and transmits the resource appeal information of each virtual machine to a Virtual Infrastructure Manager (VIM), wherein the resource appeal information comprises bandwidth resource appeal information, and the VNFD is a configuration template for describing deployment and operation behaviors of the VNF;
and the VIM deploys the virtual machines according to the resource appeal information of the virtual machines, so that the sum of the bandwidth resource appeal of the virtual machines deployed on each node is smaller than the upper limit of the bandwidth.
2. The method of claim 1, wherein the VIM deploying each virtual machine according to the resource appeal information of each virtual machine comprises:
the VIM deploys each virtual machine on each node in a bandwidth balancing mode according to the bandwidth resource appeal information of each virtual machine, so that the difference value of the sum of the bandwidth resource appeal of the virtual machines deployed on each node is smaller than a threshold value;
or the VIM deploys each virtual machine on each node in a node compact deployment manner according to the bandwidth resource appeal of each virtual machine, so that only when the bandwidth resource on one node cannot continuously accommodate the virtual machine, the virtual machine is continuously deployed on the next node of the node.
3. The method of claim 1, further comprising:
when a virtual machine is added, the VIM deploys the added virtual machine on a node meeting the bandwidth resource appeal of the added virtual machine according to the bandwidth resource occupation condition of each node.
4. The method of claim 1, further comprising:
when virtual machine migration occurs, the VIM automatically deploys the virtual machines to be migrated on the nodes meeting the bandwidth resource requirements of the virtual machines to be migrated according to the bandwidth resource occupation condition of each node;
alternatively, the first and second electrodes may be,
when the virtual machine is migrated, the VIM judges whether a node designated by a user meets the bandwidth resource demand of the virtual machine needing migration; if yes, migrating the virtual machine needing to be migrated to the node specified by the user; if not, the migration is not performed.
5. The method of claim 1, further comprising:
when a node with a virtual machine is in fault, for each virtual machine on the node in fault, the VIM reconstructs the virtual machine on the node meeting the bandwidth resource demand of the virtual machine according to the bandwidth resource occupation condition of each node.
6. The method of any of claims 1-5, wherein the resource appeal further includes at least one of: CPU resource appeal, memory resource appeal and hard disk resource appeal.
7. A system for deploying virtual machines in a network functions virtualization, NFV, the system comprising a virtual network functions manager, VNFM, and a virtualization infrastructure manager, VIM, wherein:
the VNFM is used for acquiring resource appeal information of each virtual machine through the VNFD, and transmitting the resource appeal information of each virtual machine to the VIM, wherein the resource appeal information comprises bandwidth resource appeal information, and the VNFD is a configuration template for describing deployment and operation behaviors of the VNF;
the VIM is used for deploying the virtual machines according to the resource appeal information of the virtual machines transmitted by the VNFM, so that the sum of the bandwidth resource appeal of the virtual machines deployed on each node is smaller than the upper limit of the bandwidth.
8. The system of claim 7, wherein the VIM is specifically configured to:
according to the bandwidth resource appeal of each virtual machine, deploying each virtual machine on each node in a bandwidth balancing mode, and enabling the difference value of the sum of the bandwidth resource appeal of the virtual machines deployed on each node to be smaller than a threshold value;
or according to the bandwidth resource appeal of each virtual machine, each virtual machine is deployed on each node in a node compact deployment mode, so that the virtual machine is continuously deployed on the next node of the node only when the bandwidth resource on one node cannot continuously accommodate the virtual machine.
9. The system of claim 7, wherein the VIM is further configured to:
when the virtual machines are added, the added virtual machines are deployed on the nodes meeting the bandwidth resource requirements of the added virtual machines according to the bandwidth resource occupation condition of each node.
10. The system of claim 7, wherein the VIM is further configured to:
when the virtual machine is migrated, automatically deploying the virtual machine to be migrated on a node meeting the bandwidth resource appeal of the virtual machine to be migrated according to the bandwidth resource occupation condition of each node;
alternatively, the first and second electrodes may be,
when the virtual machine is migrated, judging whether a node specified by a user meets the bandwidth resource appeal of the virtual machine needing to be migrated; if yes, migrating the virtual machine needing to be migrated to the node specified by the user; if not, the migration is not performed.
11. The system of claim 7, wherein the VIM is further configured to:
when a node with a virtual machine is in fault, for each virtual machine on the node in fault, reconstructing the virtual machine on the node meeting the bandwidth resource appeal of the virtual machine according to the bandwidth resource occupation condition of each node.
12. The system of any of claims 7-11, wherein the resource appeal further includes at least one of: CPU resource appeal, memory resource appeal and hard disk resource appeal.
13. A computer-readable storage medium comprising instructions that, when executed on a computer, cause the computer to perform the method of any of claims 1-6.
CN201710054079.XA 2017-01-24 2017-01-24 Method and system for deploying virtual machine in NFV Active CN108345490B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710054079.XA CN108345490B (en) 2017-01-24 2017-01-24 Method and system for deploying virtual machine in NFV

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710054079.XA CN108345490B (en) 2017-01-24 2017-01-24 Method and system for deploying virtual machine in NFV

Publications (2)

Publication Number Publication Date
CN108345490A CN108345490A (en) 2018-07-31
CN108345490B true CN108345490B (en) 2021-04-09

Family

ID=62962841

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710054079.XA Active CN108345490B (en) 2017-01-24 2017-01-24 Method and system for deploying virtual machine in NFV

Country Status (1)

Country Link
CN (1) CN108345490B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111949364A (en) * 2019-05-16 2020-11-17 华为技术有限公司 Deployment method of containerized VNF and related equipment

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102446115A (en) * 2012-01-09 2012-05-09 浙江大学 Dynamic deployment method for virtual machines
CN103077079A (en) * 2012-12-28 2013-05-01 华为技术有限公司 Method and device for controlling migration of virtual machine
CN104010028A (en) * 2014-05-04 2014-08-27 华南理工大学 Dynamic virtual resource management strategy method for performance weighting under cloud platform
CN104572251A (en) * 2015-01-30 2015-04-29 中国联合网络通信集团有限公司 Virtual machine deploying method and device
CN104750541A (en) * 2015-04-22 2015-07-01 成都睿峰科技有限公司 Virtual machine migration method

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8850426B2 (en) * 2009-12-13 2014-09-30 International Business Machines Corporation Managing remote deployment of a virtual machine and service request to be processed by the virtual machines based on network bandwith and storage connectivity

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102446115A (en) * 2012-01-09 2012-05-09 浙江大学 Dynamic deployment method for virtual machines
CN103077079A (en) * 2012-12-28 2013-05-01 华为技术有限公司 Method and device for controlling migration of virtual machine
CN104010028A (en) * 2014-05-04 2014-08-27 华南理工大学 Dynamic virtual resource management strategy method for performance weighting under cloud platform
CN104572251A (en) * 2015-01-30 2015-04-29 中国联合网络通信集团有限公司 Virtual machine deploying method and device
CN104750541A (en) * 2015-04-22 2015-07-01 成都睿峰科技有限公司 Virtual machine migration method

Also Published As

Publication number Publication date
CN108345490A (en) 2018-07-31

Similar Documents

Publication Publication Date Title
US11036536B2 (en) Method, apparatus, and system for deploying virtualized network function using network edge computing
US9999030B2 (en) Resource provisioning method
EP3461087B1 (en) Network-slice resource management method and apparatus
CN107924383B (en) System and method for network function virtualized resource management
EP3200393B1 (en) Method and device for virtual network function management
WO2016107418A1 (en) Allocation method, apparatus and system for cloud network communication path
US9692707B2 (en) Virtual resource object component
CN105979007B (en) Method and device for accelerating resource processing and network function virtualization system
US10644952B2 (en) VNF failover method and apparatus
WO2016165292A1 (en) Method and apparatus for realizing deployment specification configuration of virtual network function
CN106657173B (en) Service migration method, device and server in software upgrading under NFV architecture
US20170373931A1 (en) Method for updating network service descriptor nsd and apparatus
US10972365B2 (en) Method and entities for service availability management
EP3211531B1 (en) Virtual machine start method and apparatus
CN108345490B (en) Method and system for deploying virtual machine in NFV
JP6369730B2 (en) Virtual machine starting method and apparatus
CN107534577B (en) Method and equipment for instantiating network service
CN107408058B (en) Virtual resource deployment method, device and system
CN111698112B (en) Resource management method and device for VNF (virtual network function)
CN110545193B (en) Virtual resource management method, virtual resource management equipment and server
CN111355602B (en) Resource object management method and device
US20190158354A1 (en) Resource configuration method and apparatus
CN112889247A (en) VNF service instantiation method and device
CN113098705A (en) Authorization method and device for life cycle management of network service
CN111221620A (en) Storage method, storage device and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant