CN107026802B

CN107026802B - Resource management method and equipment

Info

Publication number: CN107026802B
Application number: CN201610071143.0A
Authority: CN
Inventors: 陈立福; 徐日东; 张文革
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2016-01-30
Filing date: 2016-01-30
Publication date: 2020-12-15
Anticipated expiration: 2036-01-30
Also published as: CN107026802A; WO2017128979A1

Abstract

The embodiment of the invention provides a resource management method and equipment, wherein the method comprises the following steps: the resource management equipment acquires state parameters of a plurality of VNFs included in the E2E network, and determines a first VNF to be expanded and a first resource capacity required by expansion from the VNFs according to the state parameters; if the free resource capacity of the resource pool is smaller than the first resource capacity, the resource management device determines a first VNF to be reduced from the VNFs according to the state parameter, and performs capacity reduction processing on the first VNF to be reduced according to the first resource capacity; if the spare resource capacity of the resource pool is greater than or equal to the first resource capacity after the resource management device performs capacity reduction processing on the first VNF to be subjected to capacity reduction processing, the resource management device performs capacity expansion processing on the first VNF to be subjected to capacity expansion processing. The embodiment of the invention can ensure that the resources are configured in balance among the VNFs in the E2E network under the condition of insufficient idle resources, thereby improving the resource utilization rate and the service success rate of the E2E network.

Description

Resource management method and equipment

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a resource management method and device.

Background

The traditional telecommunication network is composed of various physical devices, different network services need different physical devices, and the telecommunication network becomes more and more large and complex along with the rapid increase of the network scale, which also brings many problems, such as slow online development of new services, complex operation and maintenance of the system, high cost, low resource utilization rate and the like.

The NFV technology can be understood as migrating The functions of each Network element used in a telecommunication Network system from a current dedicated hardware platform to a general Commercial Shelf product (COTS) server, converting each Network element in The telecommunication Network into an independent application, and flexibly deploying The independent application on a unified infrastructure platform constructed based on devices such as a standard server, a storage and a switch. The general hardware equipment such as COTS calculation/storage/network can be decomposed into a plurality of virtual resources through a virtualization technology so as to be used by various upper-layer applications, and decoupling between the applications and the hardware is realized.

For an End-To-End (English: End To End, abbreviation: E2E) network applying NFV technology, compared with the front-End network element overload flow control, the back-End network element overload flow control can cause more network resource waste, and the service success rate of the E2E network is lower; and when the front-end network element overload flow control is adopted and the load of the rear-end network element is light, the service success rate of the E2E network cannot be maximized because the network element resource with light load is not fully utilized. Although the service success rate of the E2E network can be improved by directly expanding the capacity of the network element for overload flow control, this requires that the idle resources in the resource pool are sufficient, and the resource utilization rate at this time is not high.

Disclosure of Invention

The embodiment of the invention provides a resource management method and equipment, which can ensure that resources are uniformly configured among VNFs in an E2E network under the condition of insufficient idle resources, so that the resource utilization rate and the service success rate of the E2E network are improved.

A first aspect of an embodiment of the present invention provides a resource management method applied to an E2E network, where the E2E network includes multiple VNFs and a resource management device that manages the multiple VNFs and a resource pool, and the method includes:

the resource management device obtains the state parameters of the plurality of VNFs, determines an expansion object from the plurality of VNFs according to the state parameters, records the expansion object as a first VNF to be expanded, and determines the resource capacity required for expanding the first VNF to be expanded, and records the resource capacity as the first resource capacity.

If the current free resource capacity of the resource pool does not reach the first resource capacity, the resource management device first selects a VNF for capacity reduction, which may be determining a capacity reduction object from the VNFs according to the state parameter, marking as a first VNF to be capacity reduced, and performing capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity.

If the free resource capacity of the resource pool reaches the first resource capacity after the first VNF to be scaled is scaled, the resource management device may perform capacity expansion processing on the first VNF to be scaled, so as to ensure that resources are configured in a balanced manner among VNFs in the E2E network under the condition that the free resources in the resource pool are insufficient, thereby improving the resource utilization rate and the service success rate of the E2E network.

Optionally, the state parameter includes a service success rate, the resource management device may determine a current first service success rate of the E2E network according to the service success rates of the VNFs, if the first service success rate is less than or equal to a preset first service success rate, the resource management device determines a first VNF to be expanded and a first resource capacity required for expansion from the VNFs according to the state parameter, and only when the service success rate of the E2E network cannot meet a service requirement, the resource of the resource pool is reconfigured.

Optionally, after the resource management device performs the capacity reduction processing on the first VNF to be capacity reduced, or, after the capacity expansion processing is performed on the first VNF to be subjected to capacity expansion, the state parameters of the VNFs may be obtained again, and determining a current second service success rate of the E2E network according to the service success rates of the plurality of VNFs obtained again, if the service success rate of the E2E network at this time is reduced significantly, i.e. the second service success rate is less than the first service success rate, and the difference between the first service success rate and the second service success rate is greater than or equal to a preset first value, the resource management device withdraws the capacity reduction processing on the first VNF to be capacity reduced or the capacity expansion processing on the first VNF to be capacity expanded, the capacity reduction processing and the capacity expansion processing can be backed off, and the condition that the network state of E2E is worsened after resource reconfiguration is effectively improved.

Optionally, after the resource management device performs the capacity reduction processing on the first VNF to be capacity reduced, or, after the capacity expansion processing is performed on the first VNF to be expanded, when the service success rate of the E2E network is not significantly increased, that is, the second service success rate is greater than or equal to the first service success rate, and the difference between the second service success rate and the first service success rate is less than or equal to a preset second value, or, when the service success rate of the E2E network is slightly decreased, i.e. the second service success rate is less than the first service success rate, and the difference between the first service success rate and the second service success rate is less than or equal to a preset third value, the resource management device can delay the preset time length, wait for the E2E network to enter the stable working state and then carry out resource reconfiguration, the accuracy of resource reconfiguration can be improved, and the resource utilization rate and the service success rate of the E2E network can be further improved.

Optionally, after the resource management device performs capacity reduction processing on the first VNF to be subjected to capacity reduction, and when the free resource capacity of the resource pool still does not reach the first resource capacity, the capacity reduction object may be selected again according to the state parameter, and is recorded as a second VNF to be subjected to capacity reduction, and the second VNF to be subjected to capacity reduction is performed according to the free resource capacity of the resource pool and the first resource capacity, after the second VNF to be subjected to capacity reduction, if the free resource capacity of the resource pool reaches the first resource capacity, the resource management device performs capacity expansion processing on the first VNF to be subjected to capacity reduction, that is, the VNF can be continuously selected for capacity reduction when the free resource capacity of the resource pool after one time of capacity reduction still is insufficient, until the free resource capacity of the resource pool after capacity reduction reaches the requirement.

Optionally, the state parameters include resource occupancy, service success rate, service priority, and network location, and the specific manner in which the resource management device determines the first VNF to be expanded and the first resource capacity required for expansion from the VNFs according to the state parameters may be as follows:

and taking the VNF of which the state parameter meets the preset expansion condition in the plurality of VNFs as a first VNF to be expanded, and determining a first resource capacity required for expansion according to the state parameter of the first VNF to be expanded.

Wherein, the preset expansion condition comprises one or more of the following expansion conditions:

the resource occupancy rate is greater than or equal to a preset first resource occupancy rate, the service success rate is less than or equal to a preset second service success rate, the service priority is greater than or equal to a preset first service priority, and the network location is a back end.

Optionally, the specific manner in which the resource management device determines the first VNF to be scaled from the plurality of VNFs according to the state parameter may be:

and taking the VNF of which the state parameter meets the preset capacity reduction condition in the plurality of VNFs as a first VNF to be subjected to capacity reduction.

Wherein, the preset capacity reducing conditions comprise one or more of the following capacity reducing conditions:

the resource occupancy rate is less than or equal to the preset second resource occupancy rate, the service success rate is greater than or equal to the preset third service success rate, the service priority is less than or equal to the preset second service priority, and the network position is the front end.

Optionally, the specific manner of performing, by the resource management device, the capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity may be as follows: and acquiring the running states of the VMs running on the first VNF to be reduced, including the CPU occupancy rate, the RAM occupancy rate and the like, determining the VMs to be closed according to the first resource capacity and the running states, and closing the VMs to be closed, so that the first VNF to be reduced releases partial resources to increase the free resource capacity of the resource pool.

Optionally, the specific manner of performing capacity expansion processing on the first VNF to be subjected to capacity expansion is as follows:

and creating a VM and/or a starting VM on the first VNF to be expanded, and configuring the resources with the first resource capacity to the created VM and/or the starting VM.

A second aspect of the present invention provides a resource management device, which is applied to an E2E network, where the E2E network includes multiple VNFs and the resource management device that manages the multiple VNFs and a resource pool, and the device includes:

an obtaining module, configured to obtain the state parameters of the VNFs.

And the determining module is configured to determine an expansion object from the multiple VNFs according to the state parameter, record the expansion object as a first VNF to be expanded, determine a resource capacity required for expanding the first VNF to be expanded, and record the resource capacity as the first resource capacity.

The determining module is further configured to determine, according to the state parameter, a capacity reduction object from the multiple VNFs and record the capacity reduction object as a first VNF to be reduced, when the current free resource capacity of the resource pool does not reach the first resource capacity.

And the capacity reduction module is used for carrying out capacity reduction processing on the first VNF to be subjected to capacity reduction according to the first resource capacity.

And the capacity expansion module is used for performing capacity expansion processing on the first VNF to be subjected to capacity expansion under the condition that the free resource capacity of the resource pool reaches the first resource capacity after the first VNF to be subjected to capacity expansion, so that the resources can be ensured to be configured in a balanced manner among the VNFs in the E2E network under the condition that the free resources in the resource pool are insufficient, and further the resource utilization rate and the service success rate of the E2E network are improved.

Optionally, the status parameter includes a service success rate,

the specific manner of determining, by the determining module, the first VNF to be expanded and the first resource capacity required for expansion from the multiple VNFs according to the state parameter may be:

determining a current first service success rate of the E2E network according to the service success rates of the VNFs, and determining a first VNF to be expanded and a first resource capacity required for expansion according to the state parameter when the first service success rate is less than or equal to a preset first service success rate, that is, reconfiguring resources of the resource pool only when the service success rate of the E2E network cannot meet a service requirement.

Alternatively to this, the first and second parts may,

the obtaining module is further configured to obtain the state parameters of the multiple VNFs again after the first VNF to be capacity-reduced is capacity-reduced or after the first VNF to be capacity-expanded is capacity-expanded.

The determining module is further configured to determine a current second service success rate of the E2E network according to the service success rates of the VNFs obtained again by the obtaining module.

Wherein, this equipment can also include:

and the rollback module is configured to, when the service success rate of the E2E network is significantly decreased, that is, the second service success rate is smaller than the first service success rate, and a difference between the first service success rate and the second service success rate is greater than or equal to a preset first value, withdraw the capacity reduction processing performed on the first VNF to be capacity reduced or expand the capacity of the first VNF to be capacity expanded, and may rollback the capacity reduction processing and the capacity expansion processing to effectively improve a situation that the E2E network state is degraded after resource reconfiguration.

Optionally, the apparatus may further include:

the delay module is configured to delay a preset time when the service success rate of the E2E network is not significantly increased, that is, the second service success rate is greater than or equal to the first service success rate, and a difference between the second service success rate and the first service success rate is less than or equal to a preset second value, or when the service success rate of the E2E network is slightly decreased, that is, the second service success rate is less than the first service success rate, and a difference between the first service success rate and the second service success rate is less than or equal to a preset third value, and perform resource reconfiguration after the E2E network enters a stable operating state, so that accuracy of resource reconfiguration can be improved, and resource utilization rate and service success rate of the E2E network are further improved.

Alternatively to this, the first and second parts may,

the determining module is further configured to, after the first VNF to be scaled down is scaled down, select the scaling object again according to the state parameter and mark the scaling object as a second VNF to be scaled down when the free resource capacity of the resource pool still does not reach the first resource capacity.

The capacity reduction module is further configured to reduce the capacity of the second VNF to be reduced according to the free resource capacity of the resource pool and the first resource capacity.

The capacity expansion module is further configured to, after the second VNF to be subjected to capacity reduction, perform capacity expansion processing on the first VNF to be subjected to capacity expansion if the free resource capacity of the resource pool reaches the first resource capacity, that is, continue to select VNF capacity reduction when the free resource of the resource pool after one time of capacity reduction is still insufficient until the free resource capacity of the resource pool after capacity reduction reaches the requirement.

Optionally, the state parameters include resource occupancy, service success rate, service priority, and network location, and the specific manner in which the determining module determines the first VNF to be expanded and the first resource capacity required for expansion from the multiple VNFs according to the state parameters may be as follows:

Optionally, the specific manner in which the determining module determines the first VNF to be scaled from the plurality of VNFs according to the state parameter may be:

Optionally, the capacity reduction module may specifically include:

and the obtaining submodule is used for obtaining the running state of the VM running on the first VNF to be reduced, wherein the running state comprises CPU (Central processing Unit) occupancy rate, RAM (random Access memory) occupancy rate and the like.

And the determining submodule is used for determining the VM to be shut down according to the first resource capacity and the running state.

And the shutdown submodule is used for shutting down the VM to be shutdown, so that the first VNF to be scaled down releases part of resources to increase the free resource capacity of the resource pool.

Optionally, the capacity expansion module may specifically include:

and the starting submodule is used for creating the VM and/or starting the VM on the first VNF to be expanded.

And the configuration submodule is used for configuring the resources with the first resource capacity to the created VM and/or the started VM.

A third aspect of the embodiments of the present invention provides a resource management device, which is applied to an E2E network, where the E2E network includes multiple VNFs and the resource management device that manages the multiple VNFs and a resource pool, and the device includes a processor, a network interface, and a memory that stores executable program codes, where the processor calls the executable program codes stored in the memory to obtain state parameters of the multiple VNFs through the network interface, determines, according to the state parameters, a capacity expansion object from the multiple VNFs, and records the capacity expansion object as a first VNF to be expanded, and determines a resource capacity required for expanding the first VNF to be expanded, and records the capacity as the first resource capacity.

The processor is further configured to, when the current free resource capacity of the resource pool does not reach the first resource capacity, first select a VNF for capacity reduction, where a capacity reduction object is determined from the multiple VNFs according to the state parameter, and is marked as a first VNF to be capacity reduced, and perform capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity.

The processor is further configured to perform capacity expansion processing on the first VNF to be subjected to capacity expansion when the free resource capacity of the resource pool reaches the first resource capacity after the first VNF to be subjected to capacity expansion is subjected to capacity expansion, so that resources can be guaranteed to be configured in a balanced manner among VNFs in the E2E network under the condition that the free resources in the resource pool are insufficient, and resource utilization rate and service success rate of the E2E network are improved.

Optionally, the state parameter includes a service success rate, and a specific manner in which the processor determines the first VNF to be expanded and the first resource capacity required for expansion from the VNFs according to the state parameter may be as follows:

determining a current first service success rate of the E2E network according to the service success rates of the VNFs, if the first service success rate is less than or equal to a preset first service success rate, determining a first VNF to be expanded and a first resource capacity required for expansion from the VNFs according to the state parameter, and reconfiguring resources of the resource pool only when the service success rate of the E2E network cannot meet a service requirement.

In the embodiment of the present invention, a resource management device obtains state parameters of multiple VNFs included in an E2E network, and determines, according to the state parameters, a first VNF to be expanded and a first resource capacity required for expansion from the multiple VNFs; if the current free resource capacity of the resource pool is smaller than the first resource capacity, the resource management device determines a first VNF to be reduced from the VNFs according to the state parameter, and performs capacity reduction processing on the first VNF to be reduced according to the first resource capacity to release resources into the resource pool; if the spare resource capacity of the resource pool is greater than or equal to the first resource capacity after the resource management device performs capacity reduction on the first VNF to be capacity reduced, the resource management device performs capacity expansion on the first VNF to be capacity-expanded, so that resources can be guaranteed to be configured in a balanced manner among VNFs in the E2E network under the condition that spare resources are insufficient, and resource utilization rate and service success rate of the E2E network are improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a schematic diagram of an E2E network architecture according to an embodiment of the present invention;

fig. 2 is a schematic diagram of an IMS cloud network structure according to an embodiment of the present invention;

fig. 3 is a flowchart illustrating a resource management method according to an embodiment of the present invention;

FIG. 4 is a flow chart illustrating another resource management method according to an embodiment of the present invention;

FIG. 5 is a flowchart illustrating a resource management method according to another embodiment of the present invention;

fig. 6 is a schematic structural diagram of a resource management device according to an embodiment of the present invention;

fig. 7 is a schematic structural diagram of another resource management device according to an embodiment of the present invention;

fig. 8 is a schematic structural diagram of another resource management device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Fig. 1 is a schematic diagram of an E2E network architecture according to an embodiment of the present invention. The E2E network described in this embodiment mainly includes the following functional components:

the NFV Infrastructure (NFV Infrastructure, abbreviated as NFVI) is responsible for providing virtualized resources such as Virtual Computing, Virtual Storage, and Virtual Network (Virtual Network) required for the operation of NFV, and includes bottom Layer Hardware such as COTS' Computing Hardware (Computing Hardware), Storage Hardware (Storage Hardware), and Network Hardware (Network Hardware), and a Virtualization Layer (Virtualization Layer) for virtualizing the bottom Layer Hardware.

The Virtual Network Function (abbreviated as VNF) refers to a software implementation of a Network Function (abbreviated as NF) running on the NFVI, and may be further attached with an Element Management System (abbreviated as EMS) for managing an individual VNF and its characteristics. The VNF corresponds to an entity of a Network node, i.e. a Network Element (NE).

NFV Management and Orchestration (MANO), the lifecycle Management of the physical and/or software resources responsible for Orchestration, supporting the underlying hardware virtualization, and the lifecycle Management of the VNF. The NFV MANO may also interact with an Operation Support System (OSS)/service support system (BSS) outside the NFV framework to implement NFV services according to a service request of the OSS/BSS.

The above-mentioned functional components are designed in a way that allows existing solutions, such as IP Multimedia Subsystem (IMS), to be reused, and also to interact with existing OSS/BSS that need to be connected.

Among them, NFV MANO internally defines the following functional components:

the NFV Orchestrator (NFV editor, acronym: NFVO) is responsible for lifecycle management of network services, completing network services orchestration, and NFVI resource orchestration across multiple Virtualized Infrastructure Managers (VIM).

A VNF Manager (english: VNF Manager, abbreviation: VNFM) responsible for lifecycle management of VNF instances, monitoring state parameters of VNFs, such as resource occupancy rates, service success rates, service priorities, etc., of the VNFs, and a VNFM may correspondingly manage a single VNF instance or multiple VNF instances of the same or different types, including: VNF instantiation, VNF configuration NFVI resources, VNF instance update, VNF instance scaling, VNF instance termination, etc.

The VIM, responsible for controlling and managing the NFVI's computing, storage, and network resources, is typically within an operator's infrastructure sub-domain. A VIM may be dedicated to handle certain types of NFVI resources or may manage multiple types of NFVI resources, including: orchestrate NFVI resource configuration/upgrade/deconfiguration/recycle, manage the association of virtualized resources (virtualized compute, virtualized storage, virtualized networks) with underlying hardware resources (compute hardware, storage hardware, network hardware), manage the inventory of underlying hardware resources and software resources (e.g., hypervisors).

On this basis, a functional component CTRL, which may be referred to as a resource reconfiguration control module, may also be defined inside the NFV MANO, and is mainly responsible for implementing a resource reconfiguration algorithm between VNFs, so that NFVI resources are configured equally among VNFs. The CTRL functional component may be provided separately or integrated in the VNFM or VIM.

It should be noted that the NFV MANO may be deployed on a general physical network device or a physical server; or may be deployed on a Virtual Machine (VM), for example, a VM running on a VNF, which is not limited in the embodiments of the present invention.

Taking the example that the E2E network shown in fig. 1 is the IMS cloud network shown in fig. 2, the IMS cloud network includes the following network elements: a Session Border Controller (SBC for short), a Call Session Control Function (CSCF for short), a Home Subscriber Server (HSS for short), a universal voice service Server (ATS for short), and a NFV MANO. SBC, CSCF, HSS, and ATS are VNF in the E2E network architecture shown in fig. 1, where the SBC is specifically two (SBC1 and SBC2), SBC1 and SBC2 are access side (i.e., front end) network elements, CSCF, HSS, and ATS are core side (i.e., back end) network elements, and NFV MANO is deployed on a resource management device and used to monitor state parameters of SBC1, SBC2, CSCF, HSS, and ATS, and a User Equipment (User Equipment, abbreviated as UE) accesses the IMS cloud network through SBC1 or SBC2, assuming that a total amount of resources of a resource pool of the IMS cloud network is constant.

The resource management device periodically acquires state parameters such as resource occupancy rates, service success rates, service priorities, network locations and the like of the SBC1, the SBC2, the CSCF, the HSS and the ATS, determines the service success rate of the IMS cloud network according to the service success rates of the SBC1, the SBC2, the CSCF, the HSS and the ATS, and can start resource reconfiguration of the resource pool if the service success rate of the IMS cloud network is less than a certain threshold (e.g., 95%). The resource reconfiguration process of a typical application scenario is mainly explained in terms of resource occupancy.

Specifically, if the access side network element is overloaded and the core side network element is idle, the resource occupancy rate is shown in table 1:

TABLE 1

Network element	SBC1	SBC2	CSCF	HSS	ATS
						Resource occupancy rate	100％	100％	40％	70％	50％

The resource management device may determine SBC1 and SBC2 as capacity-expansion objects according to the capacity-expansion condition described in step S302 in the previous embodiment, assuming that the capacity of resources required for the capacity expansion of SBC1 and SBC2 is 4 CPUs and 3GB of operating Memory (english: Random Access Memory, abbreviated as RAM), the free resource capacity of the resource pool is 1 CPU and 200M of RAM, it is obvious that the free resource capacity is insufficient to expand SBC1 and SBC2, the resource management device may determine CSCF as a capacity-reduction object according to the capacity-reduction condition described in step S303 in the previous embodiment, after the CSCF is reduced, if the free resource capacity of the resource pool is still insufficient, and at this time, the resource occupancy of ATS is the lowest, the resource management device further reduces ATS until the free resource capacity of the resource pool is greater than or equal to 4 CPUs and 3GB of RAM SBCs, the resource management device performs the capacity expansion of 1 and 2, and then may wait for an observation period, then, whether the service success rate of the IMS cloud network and/or the state parameters of the access side network element and the core side network element satisfy the resource reconfiguration end condition described in step S409 in the foregoing embodiment is determined, and if yes, the resource reconfiguration of the resource pool is ended; if not, the resource management device repeatedly executes the above operations until the service success rate of the IMS cloud network and/or the state parameters of the access side network element and the core side network element satisfy the resource reconfiguration end condition.

Similarly, if the access side network element is idle and the core side network element is overloaded, the resource occupancy rate is shown in table 2:

TABLE 2

Network element	SBC1	SBC2	CSCF	HSS	ATS
						Resource occupancy rate	70％	70％	80％	80％	100％

The resource management device may determine that the ATS is a capacity expansion object according to the capacity expansion condition, if the free resource capacity of the resource pool is insufficient and the ATS cannot be expanded, the resource management device may determine that the access side network elements SBC1 and SBC2 are capacity reduction objects according to the capacity reduction condition, which may be first capacity reduction SBC1 (or SBC2), and if the free resource capacity after capacity reduction is still insufficient, then capacity reduction SBC2 (or SBC1), or SBC1 and SBC2, which have low service priority, and if the free resource capacity after capacity reduction is still insufficient, then capacity reduction SBC with high service priority is performed until the free resource capacity after capacity reduction is greater than or equal to the resource capacity required for expanding the ATS and expanding the capacity of the ATS.

Similarly, if the core side network element is partially overloaded and partially idle, the resource occupancy rate is shown in table 3:

TABLE 3

Network element	SBC1	SBC2	CSCF	HSS	ATS
						Resource occupancy rate	90％	90％	80％	30％	100％

The resource management device may determine, according to the capacity expansion condition, that the ATS is a capacity expansion object, and if the free resource capacity of the resource pool is insufficient and the ATS cannot be expanded, the resource management device may determine, according to the capacity reduction condition, that the core-side network element HSS is a capacity reduction object, and if the free resource capacity after HSS is reduced is greater than or equal to the resource capacity required for expanding the capacity of the ATS, the ATS is expanded.

Similarly, if the resource occupancy rates of the access-side network element and the core-side network element are both high and there are overloaded network elements, the resource occupancy rates are shown in table 4:

TABLE 4

Network element	SBC1	SBC2	CSCF	HSS	ATS
						Resource occupancy rate	100％	100％	90％	90％	100％

The resource management device may determine, according to the capacity expansion condition, that the network element ATS overloaded on the core side is a capacity expansion object, and if the free resource capacity of the resource pool is insufficient, the capacity expansion of the ATS cannot be performed, because the resource occupancy rates of other network elements on the core side are also high at this time, the resource management device may determine, according to the capacity reduction condition, that the network elements SBC1 and SBC2 on the access side are capacity reduction objects, which may be, first, capacity reduction SBC1 (or SBC2), and, if the free resource capacity after capacity reduction is still insufficient, then capacity reduction SBC2 (or SBC1), or, which may be, first, low service priority SBC1 and SBC2, and, if the free resource capacity after capacity reduction is still insufficient, then capacity reduction SBC with high service priority is performed until the free resource capacity after capacity reduction is greater than or equal to the resource capacity required for capacity expansion of the ATS, and capacity expansion of the ATS is performed.

Similarly, if the network elements with high service priority are all in overload state or have high resource occupancy, assuming that the service priority of the HSS is the lowest, the resource occupancy is as shown in table 5:

TABLE 5

Network element	SBC1	SBC2	CSCF	HSS	ATS
						Resource occupancy rate	90％	90％	90％	90％	100％

The resource management device may determine that the ATS is an expansion object according to the expansion condition, if the free resource capacity of the resource pool is insufficient and the ATS cannot be expanded, because the service priority of the HSS is the lowest, even if the resource occupancy of the HSS is high at this time, the resource management device may determine that the core-side network element HSS is a contraction object according to the contraction condition, and if the free resource capacity after HSS contraction is greater than or equal to the resource capacity required for expanding the ATS, the ATS is expanded.

Please refer to fig. 3, which is a flowchart illustrating a resource management method provided by the E2E network architecture shown in fig. 1 according to an embodiment of the present invention. The E2E network includes multiple VNFs and a resource management device that manages the multiple VNFs and resource pools, and the resource management method described in this embodiment includes:

s301, the resource management device obtains status parameters of a plurality of VNFs.

The NFV MANO is deployed on the resource management device, and the status parameter may include a resource occupancy (i.e., load) of the VNF. One or more VMs are deployed on the VNF, the VNF can complete corresponding network services through the one or more VMs, and the VNF calculates the resource occupancy rate of the VNF according to the quantity of the deployed VMs, the CPU occupancy rate and the RAM occupancy rate of each VM, the registration/session resource occupancy rate of the VNF and other parameters; the status parameters may also include the amount of traffic entered, the success rate of the traffic, and the priority of the traffic. The traffic priority may be stored in the resource management device, and set when the resource management device allocates traffic for the plurality of VNFs. The state parameter may further include a network location, the resource management device may determine, according to a sequence of the service passing through the VNFs, a location of each VNF in the E2E network, and may be divided into a front-end (access-side) VNF and a back-end (core-side) VNF, and set a sequence identifier for each VNF, the resource management device stores the sequence identifier, and the location of each VNF in the E2E network may be determined according to the sequence identifier.

Specifically, the resource management device may periodically send a status parameter acquisition request to the VNFs, and receive status parameters, such as resource occupancy, service success rate, service priority, and network location, reported by the VNFs in response to the status parameter acquisition request, and the resource management device may also receive the status parameters reported by the VNFs periodically, where a period may specifically be 10 seconds (english: s), 1 minute (english: min), and the like.

Meanwhile, the resource management device may also periodically obtain the resource occupation condition of the resource pool according to the CPU occupancy, the RAM occupancy, and the like of the underlying hardware in the E2E network resource pool (i.e., NFVI resource) to determine the free resource capacity in the resource pool, where the period may specifically be 5s, 1min, and the like, and the period in which the resource management device obtains the state parameters of the VNFs may be the same as or different from the period in which the free resource capacity in the resource pool is obtained, and the embodiment of the present invention does not limit the specific values of the two periods.

S302, the resource management device determines, according to the state parameter, a first VNF to be expanded and a first resource capacity required for expansion from the VNFs.

Specifically, the resource management device determines a first service success rate of the E2E network according to the service success rates of the VNFs, where the first service success rate may be a product of the service success rates of the VNFs, and if the first service success rate is less than or equal to a preset first service success rate (e.g., 90%), the resource management device determines to start resource reconfiguration of the resource pool by the current flow, and then uses, as an expansion object (denoted as a first VNF to be expanded), a VNF of the VNFs whose state parameter meets a preset expansion condition.

The preset expansion condition may include one or more of the following expansion conditions: the resource occupancy rate is greater than or equal to a preset first resource occupancy rate (for example, 90%), the service success rate is less than or equal to a preset second service success rate (for example, 95%), the service priority is greater than or equal to a preset first service priority (namely, the service priority is higher), and the network location is the back end. If there are multiple VNFs with the same network location, the VNF with the higher service priority is preferentially expanded, for example, if there are multiple VNFs with a service success rate less than or equal to the preset second service success rate and/or a resource occupancy rate greater than or equal to the preset first resource occupancy rate, the VNF with the higher service priority is preferentially expanded, and if there are multiple VNFs with the higher service priority, the VNF with the higher service priority is preferentially expanded.

Further, the resource management device determines, according to the state parameter of the first VNF to be expanded, a resource capacity (denoted as a first resource capacity) required for expansion, and may specifically determine, according to at least one of a resource occupancy rate, a service success rate, a service priority, and a network location, the first resource capacity required for expansion of the first VNF to be expanded, for example, the higher the resource occupancy rate is, the lower the service success rate is, the higher the service priority is, the VNF located at a rear end of the expansion is, and the more the first resource capacity is required for expansion thereof. Generally, the first resource capacity may be a capacity of resources such as CPU, RAM, and the like required by running related software on one VM and VNF, that is, the resource management device uses one VM as a capacity expansion granularity during capacity expansion.

In some possible embodiments, the resource management device may also determine to start resource reconfiguration of the resource pool by the current process when the number of VNFs in the E2E network in the overload flow control state (that is, the resource occupancy rate is 100%) or the resource occupancy rate exceeds a certain threshold (for example, 97%) exceeds a preset number (for example, 3). The resource management device may also determine to start resource reconfiguration of the resource pool by the current process when the number of VNFs in the E2E network in the overload flow control state or the resource occupancy exceeds a certain threshold exceeds a preset number, and the service success rate of the E2E network is less than or equal to a certain threshold. That is, in the embodiment of the present invention, the resource management device may determine whether the resource reconfiguration of the resource pool by the current process needs to be started according to at least one of the service success rate of the E2E network and the resource occupancy rate of each VNF.

S303, when the free resource capacity of the resource pool is smaller than the first resource capacity, the resource management device determines a first VNF to be reduced from the VNFs according to the state parameter, and performs a reduction process on the first VNF to be reduced according to the first resource capacity.

Specifically, if the current free resource capacity of the resource pool does not reach the first resource capacity required for capacity expansion of the first VNF to be subjected to capacity expansion, and the resource management device cannot perform capacity expansion on the first VNF to be subjected to capacity expansion under the condition that the resource pool cannot perform capacity expansion, the resource management device may use, as a capacity reduction object (to be recorded as the first VNF to be subjected to capacity reduction), the VNF whose state parameter satisfies a preset capacity reduction condition among the plurality of VNFs.

Wherein, the preset capacity reducing conditions comprise one or more of the following capacity reducing conditions: the resource occupancy rate is less than or equal to a preset second resource occupancy rate (for example, 75%), the service success rate is greater than or equal to a preset third service success rate (for example, 97%), the service priority is less than or equal to a preset second service priority (namely, the service priority is lower), and the network location is the front end. And if multiple VNFs with the same network location exist, preferentially performing capacity reduction on the VNF with the low service priority, for example, if multiple VNFs with the service success rate larger than or equal to the preset third service success rate and/or with the resource occupancy rate smaller than or equal to the preset second resource occupancy rate exist, preferentially performing capacity reduction on a front-end VNF, and if multiple VNFs exist at the front end, preferentially performing capacity reduction on the VNF with the lowest service priority.

Further, the resource management device performs capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity required for capacity expansion, so that the first VNF to be capacity reduced releases a part of resources, and the specific manner of the capacity reduction processing may be: the resource management device obtains the running states of the VMs running on the first VNF to be reduced, including the number of VMs, the CPU occupancy and the RAM occupancy of each VM, and determines the VM to be closed that can release the resource according to the first resource capacity and the running state, and closes the VM to be closed, so that the first VNF to be reduced releases part of the resource to increase the free resource capacity of the resource pool. Generally, the resource management device may use one VM as the granularity of capacity reduction during capacity reduction.

S304, when the free resource capacity of the resource pool is greater than or equal to the first resource capacity after the resource management device performs capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity, the resource management device performs capacity expansion processing on the first VNF to be capacity expanded.

Specifically, after the first VNF to be capacity-reduced is capacity-reduced, if the current free resource capacity of the resource pool reaches the first resource capacity required for capacity expansion, the resource management device may perform capacity expansion processing on the first VNF to be capacity-expanded, where the specific manner of capacity expansion processing may be: if the first VNF to be expanded does not have a created but not started VM, the resource management device creates one or more VMs on the first VNF to be expanded, configures the resource of the first resource capacity to the created one or more VMs, and starts the created one or more VMs; if a created but un-started VM exists on the first VNF to be expanded, the resource management device may start the created but un-started VM, and configure the resources with the first resource capacity to the created but un-started VM. Meanwhile, when the VM is started, related software such as business and monitoring software on the VNF is also started.

In some possible embodiments, if the free resource capacity of the resource pool still does not reach the first resource capacity required for the expansion after the first VNF to be reduced reduces partial resources, the resource management device may obtain the status parameters of the plurality of VNFs again, and determine a reduction object (denoted as a second VNF to be reduced) from the plurality of VNFs according to the obtained status parameters of the plurality of VNFs again. Of course, if the state of the E2E network (including the service success rate, the resource occupancy of each VNF, etc.) does not change greatly after the first VNF to be reduced is reduced, the resource management device may also directly determine the second VNF to be reduced by using the state parameters of the VNFs obtained in step S301.

Further, the resource management device determines, according to the free resource capacity of the resource pool and the first resource capacity, the resource capacity that needs to be obtained by the capacity reduction again, and performs the capacity reduction processing on the second VNF to be subjected to the capacity reduction processing, and if the free resource capacity of the resource pool reaches the first resource capacity required for capacity expansion after the second VNF to be subjected to the capacity reduction, the resource management device performs the capacity expansion processing on the first VNF to be subjected to the capacity expansion processing.

In the embodiment of the present invention, a resource management device obtains state parameters of multiple VNFs included in an E2E network, and determines a first VNF to be expanded and a first resource capacity required for expansion from the multiple VNFs according to the state parameters when a service success rate of the E2E network is lower than a preset value; if the current free resource capacity of the resource pool is smaller than the first resource capacity, the resource management device determines a first VNF to be reduced from the VNFs according to the state parameter, and performs capacity reduction processing on the first VNF to be reduced according to the first resource capacity to release resources into the resource pool until the free resource capacity of the resource pool is greater than or equal to the first resource capacity, and the resource management device performs capacity expansion processing on the first VNF to be reduced, so that it can be ensured that the resources are configured in a balanced manner among VNFs in the E2E network when the free resources in the resource pool are insufficient, and further, the resource utilization rate and the service success rate of the E2E network are improved.

Please refer to fig. 4, which is a flowchart illustrating another resource management method provided by the E2E network architecture shown in fig. 1 according to an embodiment of the present invention. The E2E network includes multiple VNFs and a resource management device that manages the multiple VNFs and resource pools, and the resource management method described in this embodiment includes:

steps S401 to S403 are the same as steps S301 to S303 in the previous embodiment, and are not described again in this embodiment of the present invention.

S404, the resource management device obtains the status parameters of the VNFs again.

S405, the resource management device determines a second service success rate of the E2E network according to the service success rates of the VNFs obtained again.

Specifically, after the first VNF to be reduced is reduced, because load balancing between the VNFs needs a certain time, the resource management device may wait for a duration of an observation period, where the observation period may be a period for obtaining the state parameters of the VNFs, for example, 10s, and when the waiting duration reaches the observation period, the resource management device obtains the state parameters of the VNFs again, and determines a second service success rate of the E2E network at this time according to the service success rates of the VNFs obtained again, where the second service success rate may be a product of the service success rates of the VNFs.

S406, the resource management device determines whether the second service success rate is less than the first service success rate of the E2E network, and a difference between the first service success rate and the second service success rate is greater than or equal to a preset first value, if yes, then step S407 is executed; if not, step S408 and step S409 are executed.

Specifically, the resource management device determines whether the service success rate of the E2E network is significantly decreased after the first VNF to be reduced is reduced by comparing the second service success rate with the first service success rate before the first VNF to be reduced is reduced, that is, whether a difference between the first service success rate and the second service success rate is greater than or equal to a preset first value (for example, 3%), and if the service success rate of the E2E network is significantly decreased, step S407 is executed; if the service success rate of the E2E network does not drop or rise significantly, step S408 and step S409 are performed.

S407, the resource management device performs a rollback operation on the first VNF to be reduced.

Specifically, after the first VNF to be reduced is reduced, if the service success rate of the E2E network is significantly reduced, the resource management device reconfigures the resource released by the first VNF to be reduced to the first VNF to be reduced, starts the VM to be shut down, and cancels the reduction processing on the first VNF to be reduced, so that the E2E network is restored to the state before the first VNF to be reduced as much as possible, so as to weaken the first VNF to be reduced and cause the significant reduction of the service success rate of the E2E network.

S408, when the resource management device performs capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity and the free resource capacity of the resource pool is greater than or equal to the first resource capacity, the resource management device performs capacity expansion processing on the first VNF to be capacity expanded.

S409, when the second service success rate is greater than or equal to the first service success rate, and a difference between the second service success rate and the first service success rate is less than or equal to a preset second value, or the second service success rate is less than the first service success rate, and a difference between the first service success rate and the second service success rate is less than or equal to a preset third value, the resource management device delays for a preset time, and determines a VNF to be expanded and a resource capacity required for expansion from the VNFs without depending on state parameters of the VNFs in the delay process.

Specifically, the resource management device determines whether the service success rate of the E2E network has a significant increase or a small decrease by comparing the second service success rate with the first service success rate, and if the second service success rate is greater than or equal to the first service success rate and the difference between the second service success rate and the first service success rate is less than or equal to a preset second value (e.g., 5%), the resource management device determines that the service success rate of the E2E network has no significant increase. If the second service success rate is less than the first service success rate, and the difference between the first service success rate and the second service success rate is less than or equal to a preset third value (for example, 2%), the resource management device determines that the service success rate of the E2E network decreases slightly.

Further, when the resource management device determines that the service success rate of the E2E network does not significantly increase or has a small decrease, the resource management device may delay a preset time, and during the delay, it does not determine the VNF to be expanded and the resource capacity required for expansion from the VNFs according to the state parameters of the VNFs, that is, suspend the capacity reduction and expansion of the VNFs, where the preset time may be three times the period for obtaining the state parameters of the VNFs.

Further, the resource management device may delay the preset time length when it is determined that the service success rate of the E2E network does not significantly increase or slightly decreases after performing the capacity reduction and expansion processing for a plurality of consecutive times (e.g., 3 times), and during the delay time, the VNF to be expanded and the resource capacity required for capacity expansion are no longer determined from the VNFs according to the state parameters of the VNFs, so as to prevent the resource management device from suspending the capacity reduction and expansion of the VNF due to a certain misjudgment.

It should be noted that the resource management device may still periodically obtain the status parameters of the VNFs and the free resource capacity of the resource pool during the delay.

In some possible embodiments, steps S404 to S407 may also be performed after step S408, where after the resource management device determines that the first VNF to be expanded is expanded by comparing the second service success rate with the first service success rate before the first VNF to be expanded is expanded, if the service success rate of the E2E network is significantly reduced, the resource management device performs a rollback operation on the first VNF to be expanded, that is, releases the resource of the first resource capacity configured to the first VNF to be expanded, so that the E2E network is restored to a state before the first VNF to be expanded is expanded, so as to weaken the significant reduction of the service success rate of the E2E network caused by the expansion of the first VNF to be expanded.

In some possible embodiments, steps S404 to S407 may be performed after step S403, and also after step S408, that is, after the resource management device performs the capacity reduction processing on the first VNF to be subjected to capacity expansion processing, the resource management device compares the service success rate of the E2E network after the capacity reduction processing or the capacity expansion processing with the service success rate before the capacity reduction processing or the capacity expansion processing, and determines whether the service success rate of the E2E network is significantly reduced, if so, the resource management device performs rollback operation on the capacity reduction processing or the capacity expansion processing, namely, the resources released by the first VNF to be scaled are reconfigured to the first VNF to be scaled, or the resources of the first resource capacity configured to the first VNF to be expanded are released, so as to weaken the capacity reduction process or the capacity expansion process, which results in a significant decrease in the success rate of the E2E network service.

Further, when the second service success rate and/or the state parameters of the VNFs meet a preset resource reconfiguration end condition, the resource management device does not determine, according to the state parameters of the VNFs, a VNF to be expanded and a resource capacity required for expansion from the VNFs, so as to end resource reconfiguration of the resource pool in the current flow, where the preset resource reconfiguration end condition may specifically include one or more of the following resource reconfiguration end conditions:

1) the second service success rate of the E2E network is greater than or equal to a preset fourth service success rate (for example, 98%), that is, the service success rate of the E2E network is obviously improved after the VNF is subjected to capacity reduction and then capacity expansion, or the service success rate of the E2E network meets the requirement.

2) The resource occupancy rates of the VNFs are all less than or equal to a preset third resource occupancy rate (e.g., 90%), that is, the E2E network is in a load balancing state after the VNFs are subjected to capacity reduction and then capacity expansion.

3) The overload flow control of the front-end VNF in the VNFs (for example, the resource occupancy reaches 100%) and the absolute value of the resource occupancy difference between any two other VNFs is less than or equal to a preset fourth value (for example, 3%), that is, after the VNFs are subjected to capacity reduction and capacity expansion, although the front-end VNF is in the overload flow control state in the E2E network, the load balancing state between the other VNFs has already been reached, at this time, the E2E network may also be considered to be in the load balancing state.

4) The plurality of VNFs have no releasable resources, and the resource management device may set a minimum resource capacity occupied by each VNF, and if the resource capacity occupied by each VNF is the minimum resource capacity, the VNFs cannot be scaled down again even if the resource occupancy rate of the VNF is low.

5) The service success rate of the E2E network after continuous multiple (e.g., 3) occurrences of capacity reduction and capacity expansion is not significantly increased compared to the service success rate before capacity reduction and capacity expansion, and/or the E2E network after continuous multiple (e.g., 3) occurrences of capacity reduction and capacity expansion still does not reach a load balancing state, which may be specifically described in 2) and 3).

It should be noted that, after the resource reconfiguration of the resource pool by the current process is finished, the resource management device may circularly execute step S401 to step S409.

In the embodiment of the present invention, a resource management device obtains state parameters of multiple VNFs included in an E2E network, and determines a first VNF to be expanded and a first resource capacity required for expansion from the multiple VNFs according to the state parameters when a service success rate of the E2E network is lower than a preset value; if the current free resource capacity of the resource pool is smaller than the first resource capacity, the resource management device determines a first VNF to be reduced from the VNFs according to the state parameter, and performs capacity reduction processing on the first VNF to be reduced according to the first resource capacity to release resources into the resource pool; the resource management device determines a second service success rate of the E2E network according to the service success rates of the VNFs obtained again, determines whether the service success rate of the E2E network is significantly reduced by comparing the second service success rate with a first service success rate of the E2E network obtained before capacity reduction, and if so, backs back to capacity reduction processing of the first VNF to be subjected to capacity reduction. If not, when the free resource capacity of the resource pool is greater than or equal to the first resource capacity, performing capacity expansion processing on the first VNF to be subjected to capacity expansion, so that the resources can be guaranteed to be configured in balance among the VNFs in the E2E network under the condition that the free resources in the resource pool are insufficient, the resource utilization rate and the service success rate of the E2E network are further improved, and the capacity reduction processing and the capacity expansion processing can be backed off, so that the condition that the state of the E2E network is poor after resource reconfiguration is effectively improved. The resource management device may also determine whether the service success rate of the E2E network has an obvious increase or a small decrease by comparing the second service success rate with the first service success rate of the E2E network obtained before the capacity reduction, and if so, determine the VNF to be expanded after delaying for a period of time, that is, perform resource reconfiguration after waiting for the E2E network to enter a stable working state, so as to improve the accuracy of resource reconfiguration, and further improve the resource utilization rate and the service success rate of the E2E network.

Please refer to fig. 5, which is a flowchart illustrating a resource management method provided by the E2E network architecture shown in fig. 1 according to an embodiment of the present invention. The E2E network includes multiple VNFs and a resource management device that manages the multiple VNFs and resource pools, and the resource management method described in this embodiment includes:

steps S501 to S503 are the same as steps S301 to S303 in the previous embodiment, and are not described again in this embodiment of the present invention.

S504, the resource management device obtains the status parameters of the VNFs again.

And S505, the resource management device determines a second VNF to be expanded and a second resource capacity required by expansion from the VNFs according to the obtained state parameters of the VNFs again.

Specifically, after the resource management device performs capacity reduction processing on the first VNF to be capacity reduced, to prevent a state of the E2E network (for example, a service success rate of the E2E network, a resource occupancy rate of each VNF, and the like) from changing greatly, the state parameters of the VNFs may be obtained again, a capacity expansion object (denoted as a second VNF to be capacity expanded) may be determined again from the VNFs according to the obtained state parameters of the VNFs, a VNF of which the state parameters satisfy a preset capacity expansion condition in the VNFs may be used as the second VNF to be capacity expanded, and a resource capacity (denoted as a second resource capacity) required for capacity expansion is determined according to the state parameters of the second VNF to be capacity expanded.

The preset capacity expansion condition may refer to the related description of step S302 in the foregoing embodiment, and is not described herein again.

It may be understood that the second VNF to be expanded and the first VNF to be expanded may be the same VNF or different VNFs.

S506, when the resource management device performs capacity reduction processing on the first VNF to be subjected to capacity reduction processing according to the first resource capacity and the free resource capacity of the resource pool is greater than or equal to the second resource capacity, the resource management device performs capacity expansion processing on the second VNF to be subjected to capacity expansion processing and does not perform capacity expansion processing on the first VNF to be subjected to capacity expansion processing.

Specifically, after the first VNF to be capacity-reduced is capacity-reduced, if the current free resource capacity of the resource pool reaches the second resource capacity required for capacity expansion of the second VNF to be capacity-expanded, the resource management device may perform capacity expansion processing on the second VNF to be capacity-expanded. If the second VNF to be expanded is different from the first VNF to be expanded, the resource management device does not perform expansion processing on the first VNF to be expanded because the expansion object is the second VNF to be expanded at this time.

The specific manner of the capacity expansion processing may refer to the related description of step S304 in the foregoing embodiment, and is not described herein again.

In the embodiment of the present invention, a resource management device obtains state parameters of multiple VNFs included in an E2E network, and determines a first VNF to be expanded and a first resource capacity required for expansion from the multiple VNFs according to the state parameters when a service success rate of the E2E network is lower than a preset value; if the current free resource capacity of the resource pool is smaller than the first resource capacity, the resource management device determines a first VNF to be reduced from the VNFs according to the state parameter, and performs capacity reduction processing on the first VNF to be reduced according to the first resource capacity to release resources into the resource pool; the resource management device acquires the state parameters of the VNFs again, and re-determines the capacity expansion object (marked as a second VNF to be subjected to capacity expansion) and the second resource capacity required for capacity expansion according to the re-acquired state parameters, and if the free resource capacity of the resource pool is greater than or equal to the second resource capacity, the resource management device performs capacity expansion processing on the second VNF to be subjected to capacity expansion, and re-determines the capacity expansion object after capacity expansion, so that the situation that the selected capacity expansion object is inaccurate when the E2E network state is greatly changed due to capacity reduction can be effectively prevented, the accuracy of capacity reduction and capacity expansion is improved, and further the resource utilization rate and the service success rate of the E2E network can be improved.

Fig. 6 is a schematic structural diagram of a resource management device according to an embodiment of the present invention. The resource management device described in this embodiment is applied to an E2E network, where the E2E network includes a plurality of VNFs and the resource management device that manages the VNFs and resource pools, and includes:

an obtaining module 601, configured to obtain the state parameters of the VNFs.

A determining module 602, configured to determine, according to the state parameter, a first VNF to be expanded and a first resource capacity required for expansion from the multiple VNFs.

The state parameters include resource occupancy rate, service success rate, service priority and network location.

Specifically, the determining module 602 may use, as the first VNF to be expanded, the VNF of which the state parameter meets the preset expansion condition, and determine the first resource capacity required for expansion according to the state parameter of the first VNF to be expanded.

The preset expansion condition includes one or more of the following expansion conditions.

The determining module 602 is further configured to determine, according to the state parameter, a first VNF to be scaled down from the plurality of VNFs when the free resource capacity of the resource pool is smaller than the first resource capacity.

Specifically, the determining module 602 may use, as the first VNF to be reduced, a VNF of the plurality of VNFs whose state parameter satisfies a preset reduction condition.

A capacity reduction module 603, configured to perform capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity.

Specifically, the capacity reduction module 603 may include:

the obtaining sub-module 6030 is configured to obtain an operation state of the VM running on the first VNF to be scaled.

A determining submodule 6031 configured to determine, according to the first resource capacity and the operating state, that the VM is to be shut down.

Shutdown submodule 6032 configured to shutdown the VM to be shutdown.

The capacity expansion module 604 is configured to, after the capacity reduction module performs capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity, perform capacity expansion processing on the first VNF to be capacity expanded when the free resource capacity of the resource pool is greater than or equal to the first resource capacity.

Specifically, the capacity expansion module 604 may include:

a start module 6040, configured to create and/or start a VM on the first VNF to be expanded;

a configuration submodule 6041 configured to configure resources of the first resource capacity to the created VM and/or the started VM.

In some possible embodiments, the status parameters include a service success rate,

the determining module 602 is further configured to determine a first service success rate of the E2E network according to the service success rates of the VNFs.

The specific manner in which the determining module 602 determines the first VNF to be expanded and the first resource capacity required for expansion from the multiple VNFs according to the state parameter is as follows:

when the first service success rate of the E2E network is less than or equal to a preset first service success rate, determining a first VNF to be expanded and a first resource capacity required for expansion from the VNFs according to the state parameter.

In some of the possible embodiments of the present invention,

the obtaining module 601 is further configured to obtain the status parameters of the VNFs again.

The determining module 602 is further configured to determine a second service success rate of the E2E network according to the service success rates of the VNFs obtained again by the obtaining module.

Wherein, the resource management device further comprises:

a rollback module 605, configured to perform rollback operation on the capacity reduction processing of the first VNF to be capacity reduced or the capacity expansion processing of the first VNF to be capacity expanded when the second service success rate is less than the first service success rate and a difference between the first service success rate and the second service success rate is greater than or equal to a preset first value.

A delay module 606, configured to delay a preset duration when the second service success rate is greater than or equal to the first service success rate, and a difference between the second service success rate and the first service success rate is less than or equal to a preset second value, or the second service success rate is less than the first service success rate, and a difference between the first service success rate and the second service success rate is less than or equal to a preset third value.

The determining module 602 is further configured to determine, in the delay process of the delay module, a VNF to be expanded and a resource capacity required for expansion from the VNFs without using the state parameters of the VNFs.

Fig. 7 is a schematic structural diagram of another resource management device according to an embodiment of the present invention. The resource management device described in this embodiment is applied to an E2E network, where the E2E network includes multiple VNFs and the resource management device that manages the multiple VNFs and resource pools, and includes:

an obtaining module 701, configured to obtain the state parameters of the multiple VNFs.

A determining module 702, configured to determine, according to the state parameter, a first VNF to be expanded and a first resource capacity required for expansion from the multiple VNFs.

The determining module 702 is further configured to determine, according to the state parameter, a first VNF to be scaled down from the plurality of VNFs when the free resource capacity of the resource pool is smaller than the first resource capacity.

A capacity reduction module 703 is configured to perform capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity.

The capacity expansion module 704 is configured to, after the capacity reduction module performs capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity, perform capacity expansion processing on the first VNF to be capacity expanded when the free resource capacity of the resource pool is greater than or equal to the first resource capacity.

In some of the possible embodiments of the present invention,

the determining module 702 is further configured to determine, according to the state parameter, a second VNF to be reduced from the multiple VNFs, when the free resource capacity of the resource pool is smaller than the first resource capacity after the capacity reduction module performs capacity reduction processing on the first VNF to be reduced according to the first resource capacity.

The capacity reduction module 703 is further configured to perform capacity reduction processing on the second VNF to be subjected to capacity reduction according to the free resource capacity of the resource pool and the first resource capacity.

The capacity expansion module 704 is further configured to, after the capacity reduction module performs capacity reduction processing on the second VNF to be subjected to capacity reduction processing according to the free resource capacity of the resource pool and the first resource capacity, perform capacity expansion processing on the first VNF to be subjected to capacity expansion processing when the free resource capacity of the resource pool is greater than or equal to the first resource capacity.

In some of the possible embodiments of the present invention,

the obtaining module 701 is further configured to obtain the state parameters of the VNFs again.

The determining module 702 is further configured to determine, according to the state parameters of the VNFs obtained again by the obtaining module, a second VNF to be expanded and a second resource capacity required for expansion from the VNFs.

The capacity expansion module 704 is further configured to, when the capacity reduction module performs capacity reduction processing on the first VNF to be subjected to capacity reduction processing according to the first resource capacity and the free resource capacity of the resource pool is greater than or equal to the second resource capacity, perform capacity expansion processing on the second VNF to be subjected to capacity expansion processing, and not perform capacity expansion processing on the first VNF to be subjected to capacity expansion processing.

In the embodiment of the present invention, a resource management device obtains state parameters of multiple VNFs included in an E2E network, and determines a first VNF to be expanded and a first resource capacity required for expansion from the multiple VNFs according to the state parameters when a service success rate of the E2E network is lower than a preset value; if the current free resource capacity of the resource pool is smaller than the first resource capacity, the resource management device determines a first VNF to be reduced from the VNFs according to the state parameter, and performs capacity reduction processing on the first VNF to be reduced according to the first resource capacity to release resources into the resource pool, and if the free resource capacity of the resource pool is still insufficient after capacity reduction, continues to select the VNF for capacity reduction; the resource management device acquires the state parameters of the VNFs again, and re-determines the capacity expansion object (marked as a second VNF to be subjected to capacity expansion) and the second resource capacity required for capacity expansion according to the re-acquired state parameters, and if the free resource capacity of the resource pool is greater than or equal to the second resource capacity, the resource management device performs capacity expansion processing on the second VNF to be subjected to capacity expansion, and re-determines the capacity expansion object after capacity expansion, so that the situation that the selected capacity expansion object is inaccurate when the E2E network state is greatly changed due to capacity reduction can be effectively prevented, the accuracy of capacity reduction and capacity expansion is improved, and further the resource utilization rate and the service success rate of the E2E network can be improved.

Fig. 8 is a schematic structural diagram of another resource management device according to an embodiment of the present invention. The resource management device described in this embodiment is applied to an E2E network, where the E2E network includes multiple VNFs and the resource management device that manages the multiple VNFs and resource pools, and includes: a processor 801, a network interface 802 and a memory 803, the processor 801, the network interface 802 and the memory 803 being connected by a bus 804, wherein:

the network interface 802 is connected to the plurality of network devices such as VNFs. The number of the network interfaces 802 may be one or more. The type of network interface 802 described above may be a wired interface, a wireless interface, or a combination of both. The wired interface may be, for example, an ethernet interface, which may be an optical interface, an electrical interface, or a combination thereof, and the Wireless interface may be, for example, a Wireless Local Area Network (WLAN) interface, a cellular Network interface, or a combination thereof.

The Memory 803 may include a Volatile Memory (english: Volatile Memory), such as a RAM; Non-Volatile Memory (Non-Volatile Memory), such as Read-Only Memory (ROM), Flash Memory (HDD), Hard Disk (HDD) or Solid-State Drive (SSD); combinations of the above categories of memory may also be included.

The processor 801 is configured to call the program code stored in the memory 803, and perform the following operations:

the processor 801 is configured to obtain the status parameters of the VNFs through the network interface 802.

The processor 801 is further configured to determine, according to the state parameter, a first VNF to be expanded and a first resource capacity required for expansion from the VNFs.

The processor 801 is further configured to determine, according to the state parameter, a first VNF to be scaled down from the plurality of VNFs when the free resource capacity of the resource pool is smaller than the first resource capacity.

The processor 801 is further configured to perform capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity.

The processor 801 is further configured to, after performing capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity, perform capacity expansion processing on the first VNF to be capacity expanded when a free resource capacity of the resource pool is greater than or equal to the first resource capacity.

the processor 801 is further configured to determine a first service success rate of the E2E network according to the service success rates of the VNFs.

The specific manner of determining, by the processor 801, the first VNF to be expanded and the first resource capacity required for expansion from the multiple VNFs according to the state parameter is as follows:

Further, in the above-mentioned case,

the processor 801 is further configured to obtain the status parameters of the VNFs again through the network interface 802, and determine a second service success rate of the E2E network according to the service success rates of the VNFs.

The processor 801 is further configured to perform a rollback operation on the first VNF to be reduced or the capacity expansion of the first VNF to be expanded when the second service success rate is smaller than the first service success rate and a difference between the first service success rate and the second service success rate is greater than or equal to a preset first value.

Further, in the above-mentioned case,

the processor 801 is further configured to delay the preset time duration when the second service success rate is greater than or equal to the first service success rate, and a difference between the second service success rate and the first service success rate is less than or equal to a preset second value, or the second service success rate is less than the first service success rate, and a difference between the first service success rate and the second service success rate is less than or equal to a preset third value, and determine the VNFs to be expanded and the resource capacity required for expansion from the VNFs without using the state parameters of the VNFs in the delay process.

In some of the possible embodiments of the present invention,

the processor 801 is further configured to, after performing capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity, determine, according to the state parameter, a second VNF to be capacity reduced from the VNFs when the free resource capacity of the resource pool is smaller than the first resource capacity, and perform capacity reduction processing on the second VNF to be capacity reduced according to the free resource capacity of the resource pool and the first resource capacity.

The processor 801 is further configured to, after performing capacity reduction processing on the second VNF to be capacity reduced according to the free resource capacity of the resource pool and the first resource capacity, perform capacity expansion processing on the first VNF to be capacity expanded when the free resource capacity of the resource pool is greater than or equal to the first resource capacity.

In some possible embodiments, the status parameters include resource occupancy, service success rate, service priority, and network location, and the specific way for the processor 801 to determine the first VNF to be expanded and the first resource capacity required for expansion from the VNFs according to the status parameters is as follows:

and taking the VNF of which the state parameter meets the preset expansion condition in the plurality of VNFs as a first VNF to be expanded.

And determining the first resource capacity required by capacity expansion according to the state parameter of the first VNF to be subjected to capacity expansion.

Further, in the above-mentioned case,

the specific way for the processor 801 to determine the first VNF to be scaled from the plurality of VNFs according to the state parameter is as follows:

In some of the possible embodiments of the present invention,

the specific way for the processor 801 to perform the capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity is as follows:

and acquiring the running state of the VM running on the first VNF to be reduced.

And determining a VM to be shut down according to the first resource capacity and the running state, and shutting down the VM to be shut down.

In some of the possible embodiments of the present invention,

the specific manner of performing the capacity expansion processing on the first VNF to be subjected to capacity expansion by the processor 801 is as follows:

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.

The resource management method and device provided by the embodiment of the present invention are described in detail above, and a specific example is applied in the description to explain the principle and the implementation of the present invention, and the description of the above embodiment is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims

1. A resource management method applied to an end-to-end E2E network, the E2E network including a plurality of virtual network functions, VNFs, and a resource management device that manages the VNFs and resource pools, the method comprising:

the resource management device acquires state parameters of the plurality of VNFs;

the resource management equipment determines a first VNF to be expanded and a first resource capacity required by expansion from the plurality of VNFs according to the state parameters; the determining, by the resource management device, a first VNF to be expanded and a first resource capacity required for expansion from the VNFs according to the state parameter includes:

the resource management equipment takes the VNF of which the state parameter meets a preset capacity expansion condition as a first VNF to be subjected to capacity expansion;

the resource management equipment determines a first resource capacity required by capacity expansion according to the state parameter of the first VNF to be subjected to capacity expansion;

the resource occupancy rate is greater than or equal to a preset first resource occupancy rate, the service success rate is less than or equal to a preset second service success rate, the service priority is greater than or equal to a preset first service priority, and the network position is a rear end;

when the free resource capacity of the resource pool is smaller than the first resource capacity, the resource management device determines a first VNF to be reduced from the VNFs according to the state parameter, and performs capacity reduction processing on the first VNF to be reduced according to the first resource capacity; the resource management device determines a first VNF to be scaled down from the plurality of VNFs according to the state parameter, and includes:

the resource management equipment takes a VNF of the plurality of VNFs, of which the state parameter meets a preset capacity reduction condition, as a first VNF to be subjected to capacity reduction;

wherein the preset capacity reducing conditions comprise one or more of the following capacity reducing conditions:

the resource occupancy rate is less than or equal to a preset second resource occupancy rate, the service success rate is greater than or equal to a preset third service success rate, the service priority is less than or equal to a preset second service priority, and the network position is a front end;

and when the resource management device performs capacity reduction processing on the first VNF to be subjected to capacity reduction processing according to the first resource capacity and when the free resource capacity of the resource pool is greater than or equal to the first resource capacity, the resource management device performs capacity expansion processing on the first VNF to be subjected to capacity expansion processing.

2. The method according to claim 1, wherein the status parameter includes a service success rate, and the determining, by the resource management device, the first VNF to be expanded and the first resource capacity required for expansion from the VNFs according to the status parameter includes:

the resource management device determines a first service success rate of the E2E network according to the service success rates of the VNFs;

when the first service success rate of the E2E network is less than or equal to a preset first service success rate, the resource management device determines, according to the state parameter, a first VNF to be expanded and a first resource capacity required for expansion from the VNFs.

3. The method according to claim 2, wherein after the resource management device performs the capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity, the method further comprises:

the resource management device acquires the state parameters of the VNFs again;

the resource management device determines a second service success rate of the E2E network according to the service success rates of the plurality of VNFs obtained again;

and when the second service success rate is smaller than the first service success rate and the difference between the first service success rate and the second service success rate is greater than or equal to a preset first value, the resource management device performs rollback operation on the capacity reduction processing of the first VNF to be subjected to capacity expansion processing or on the first VNF to be subjected to capacity expansion processing.

4. The method according to claim 3, wherein after the resource management device performs the capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity, the method further comprises:

when the second service success rate is greater than or equal to the first service success rate, and a difference value obtained by subtracting the first service success rate from the second service success rate is smaller than or equal to a preset second value, or the second service success rate is smaller than the first service success rate, and a difference value obtained by subtracting the second service success rate from the first service success rate is smaller than or equal to a preset third value, the resource management device delays for a preset time length, and determines the VNFs to be expanded and the resource capacity required for expansion from the VNFs without depending on the state parameters of the VNFs in the delay process.

5. The method of claim 1, further comprising:

when the resource management device performs capacity reduction processing on the first VNF to be subjected to capacity reduction processing according to the first resource capacity and when the free resource capacity of the resource pool is smaller than the first resource capacity, the resource management device determines a second VNF to be subjected to capacity reduction from the VNFs according to the state parameter, and performs capacity reduction processing on the second VNF to be subjected to capacity reduction processing according to the free resource capacity of the resource pool and the first resource capacity;

and when the resource management device performs capacity expansion processing on the second VNF to be subjected to capacity expansion processing according to the free resource capacity of the resource pool and the first resource capacity and the free resource capacity of the resource pool is greater than or equal to the first resource capacity, the resource management device performs capacity expansion processing on the first VNF to be subjected to capacity expansion processing.

6. The method according to any one of claims 1 to 5, wherein the resource management device performs capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity, and the method comprises:

the resource management equipment acquires the running state of a virtual machine VM running on the first VNF to be reduced;

and the resource management equipment determines the VM to be closed according to the first resource capacity and the running state, and closes the VM to be closed.

7. The method according to claim 6, wherein the performing, by the resource management device, capacity expansion processing on the first VNF to be subjected to capacity expansion includes:

the resource management device creates a VM and/or starts a VM on the first VNF to be expanded, and configures the resources with the first resource capacity to the created VM and/or the started VM.

8. A resource management device for an end-to-end E2E network, the E2E network comprising a plurality of VNFs and the resource management device managing the plurality of VNFs and resource pools, the device comprising:

an obtaining module, configured to obtain state parameters of the VNFs;

a determining module, configured to determine, according to the state parameter, a first VNF to be expanded and a first resource capacity required for expansion from the multiple VNFs; the state parameters include resource occupancy rates, service success rates, service priorities and network locations, and the specific manner for the determining module to determine the first VNF to be expanded and the first resource capacity required for expansion from the VNFs according to the state parameters is as follows: taking the VNF of which the state parameter meets a preset expansion condition in the plurality of VNFs as a first VNF to be expanded; determining a first resource capacity required by capacity expansion according to the state parameter of the first VNF to be subjected to capacity expansion; wherein, the preset expansion condition comprises one or more of the following expansion conditions: the resource occupancy rate is greater than or equal to a preset first resource occupancy rate, the service success rate is less than or equal to a preset second service success rate, the service priority is greater than or equal to a preset first service priority, and the network position is a rear end;

the determining module is further configured to determine, according to the state parameter, a first VNF to be scaled down from the plurality of VNFs when a free resource capacity of the resource pool is smaller than the first resource capacity; the specific manner of determining, by the determining module, the first VNF to be scaled from the plurality of VNFs according to the state parameter is as follows:

taking the VNF of the plurality of VNFs, of which the state parameters meet preset capacity reduction conditions, as a first VNF to be subjected to capacity reduction;

the capacity reduction module is used for carrying out capacity reduction processing on the first VNF to be subjected to capacity reduction according to the first resource capacity;

and the capacity expansion module is configured to, after the capacity reduction module performs capacity reduction on the first VNF to be capacity reduced according to the first resource capacity, perform capacity expansion on the first VNF to be capacity expanded when the free resource capacity of the resource pool is greater than or equal to the first resource capacity.

9. The apparatus according to claim 8, wherein the status parameter includes a service success rate, and the specific manner in which the determining module determines, according to the status parameter, the first VNF to be expanded and the first resource capacity required for expansion from the multiple VNFs is:

determining a first service success rate of the E2E network according to the service success rates of the VNFs;

and when the first service success rate of the E2E network is less than or equal to a preset first service success rate, determining a first VNF to be expanded and a first resource capacity required for expansion from the VNFs according to the state parameter.

10. The apparatus of claim 9,

the obtaining module is further configured to obtain the state parameters of the VNFs again;

the determining module is further configured to determine a second service success rate of the E2E network according to the service success rates of the VNFs obtained again by the obtaining module;

wherein the apparatus further comprises:

and the rollback module is configured to perform rollback operation on the capacity reduction processing on the first VNF to be capacity reduced or the capacity expansion processing on the first VNF to be capacity expanded when the second service success rate is smaller than the first service success rate and a difference between the first service success rate and the second service success rate is greater than or equal to a preset first value.

11. The apparatus of claim 10, further comprising:

the delay module is used for delaying a preset time length when the second service success rate is greater than or equal to the first service success rate, and the difference value of the second service success rate minus the first service success rate is smaller than or equal to a preset second value, or the second service success rate is smaller than the first service success rate, and the difference value of the first service success rate minus the second service success rate is smaller than or equal to a preset third value;

the determining module is further configured to determine, during the delay process of the delay module, the VNFs to be expanded and the resource capacity required for expansion from the VNFs without using the state parameters of the VNFs.

12. The apparatus of claim 8,

the determining module is further configured to determine, according to the state parameter, a second VNF to be reduced from the VNFs when the free resource capacity of the resource pool is smaller than the first resource capacity after the capacity reducing module performs capacity reduction processing on the first VNF to be reduced according to the first resource capacity;

the capacity reduction module is further configured to perform capacity reduction processing on the second VNF to be subjected to capacity reduction according to the free resource capacity of the resource pool and the first resource capacity;

the capacity expansion module is further configured to, after the capacity reduction module performs capacity reduction processing on the second VNF to be subjected to capacity reduction processing according to the free resource capacity of the resource pool and the first resource capacity, perform capacity expansion processing on the first VNF to be subjected to capacity expansion processing when the free resource capacity of the resource pool is greater than or equal to the first resource capacity.

13. The apparatus of any one of claims 8 to 12, wherein the capacity reduction module comprises:

the obtaining submodule is used for obtaining the running state of the VM running on the first VNF to be scaled;

the determining submodule is used for determining the VM to be closed according to the first resource capacity and the running state;

and the closing submodule is used for closing the VM to be closed.

14. The apparatus of claim 13, wherein the capacity expansion module comprises:

the starting submodule is used for creating a VM (virtual machine) and/or starting the VM on the first VNF to be expanded;

15. A resource management device for use in an end-to-end E2E network, the E2E network comprising a plurality of VNFs and the resource management device managing the VNFs and resource pools, the device comprising a processor, a network interface, and a memory having executable program code stored therein, the processor being configured to invoke the executable program code stored in the memory to perform the following operations:

the processor to obtain, via the network interface, status parameters of the plurality of VNFs;

the processor is further configured to determine, according to the state parameter, a first VNF to be expanded and a first resource capacity required for expansion from the VNFs; the determining, by the processor, a first VNF to be expanded and a first resource capacity required for expansion from the VNFs according to the state parameter includes:

the processor takes the VNF of the plurality of VNFs, of which the state parameter meets a preset capacity expansion condition, as a first VNF to be subjected to capacity expansion;

the processor determines a first resource capacity required for capacity expansion according to the state parameter of the first VNF to be subjected to capacity expansion;

the processor is further configured to determine a first VNF to be reduced from the VNFs according to the state parameter when the free resource capacity of the resource pool is smaller than the first resource capacity, and perform capacity reduction processing on the first VNF to be reduced according to the first resource capacity;

the determining a first to-be-scaled VNF from the plurality of VNFs according to the state parameter includes:

the processor is further configured to, after the resource management device performs capacity reduction processing on the first VNF to be capacity reduced according to the first resource capacity, perform capacity expansion processing on the first VNF to be capacity expanded when an idle resource capacity of the resource pool is greater than or equal to the first resource capacity.