Method, system and device for realizing rapid capacity expansion and shrinkage of NFV network element on VI layer, and storage medium

Abstract

The invention discloses a method, a system and equipment for realizing rapid capacity expansion and shrinkage of an NFV network element on a VI layer, and a storage medium. The method comprises the following steps: introducing a virtual machine longitudinal dynamic increase and decrease distribution technology to realize rapid capacity expansion and shrinkage of a network element virtual machine; virtual machines in the recycle bin are reused, and waste virtual machine resources are reused; and a virtual machine cloning technology is introduced to complete rapid expansion of the network element virtual machine. Compared with an existing virtual machine creating mode, the method has the advantage that the rapid expansion of the NFV network element can be realized on the VI layer through the combination of various strategies. According to the scheme of the invention, the initial configuration parameter quantity of the network element parameters can be reduced; by using the scheme of the invention, the time required for capacity expansion and contraction of the NFV network element can be optimized from a minute level to a second level.

Description

Technical Field

[0001] The present invention relates to the field of cloud computing technology, and in particular to a method, system, device and storage medium for realizing rapid expansion and contraction of NFV network elements at the VI layer. Background Art

[0002] ETSI's NFV architecture is a virtualized network architecture widely used in the telecommunications field. Its goal is to decouple the functions of traditional network equipment from dedicated hardware and transfer them to general computing, storage and network resources. The 5GC three-layer decoupling architecture based on NFV technology has completed the full decoupling of the general hardware layer, virtualized infrastructure (VI) layer and virtualized network element function (VNF) layer, so as to meet the requirements of cloud deployment of unified architecture of different manufacturers and networks, realize the unified architecture and intensive sharing of cloud resource pools, effectively improve the autonomous controllability of cloud networks, improve the operational efficiency of cloud networks, and reduce construction and operation costs.

[0003] The 5G core network implemented based on the ETSI three-layer decoupling framework can dynamically adjust the capabilities of network elements according to network requirements very conveniently and efficiently, and support dynamic expansion and contraction of network elements. However, the existing network element expansion and contraction solutions mainly rely on the VIM module in the MANO component to add / delete network element virtual machines. Since it takes a long time to create / delete network element virtual machines, usually in minutes, the expansion and contraction efficiency of network elements is low. Therefore, it is very necessary to develop a method to achieve rapid expansion and contraction of network elements. Summary of the invention

[0004] The purpose of the present invention is to provide a method, system, device and storage medium for realizing rapid expansion and contraction of NFV network elements at the VI layer, and to realize rapid expansion and contraction of NFV network elements at the VI layer by combining multiple technologies such as vertical expansion and contraction of virtual machines, recovery of virtual machines by recycling bins, cloning virtual machines and creating new virtual machines.

[0005] To achieve the above purpose, the technical solution provided by the present invention is:

[0006] A first aspect of the present application provides a method for realizing rapid expansion and contraction of NFV network elements at the VI layer, comprising the following steps:

[0007] Introducing virtual machine vertical dynamic increase and decrease configuration technology to achieve rapid expansion and contraction of network element virtual machines;

[0008] Reuse virtual machines in the recycle bin to reuse abandoned virtual machine resources;

[0009] The virtual machine cloning technology is introduced to achieve rapid expansion of network element virtual machines.

[0010] To optimize the above technical solutions, the specific measures taken also include:

[0011] The described technology of dynamically increasing or decreasing the vertical configuration of virtual machines realizes the rapid scaling of network element virtual machines, including vertical expansion of virtual machine configuration, specifically as follows:

[0012] Through the VIM's POST / servers / {server_id} / action interface, and specifying the body as {"live_resize":{"vcpus":"<CPU increased configuration value>","rams":"<RAM increased configuration value>"}}, vertically increase the physical resources of the network element virtual machine according to the predetermined increased configuration value. The physical resources include CPU and memory.

[0013] Furthermore, after the VIM component receives the network element virtual machine expansion message, it compares the monitored CPU or memory physical resources of the physical machine where the network element virtual machine is located with the horizontal expansion configuration requirements; if the physical resources are sufficient, realize the rapid scaling of the network element virtual machine through the technology of dynamically increasing or decreasing the vertical configuration of virtual machines; if the physical resources are insufficient, reuse the virtual machines in the recycling station to realize the reuse of abandoned virtual machine resources.

[0014] The described reuse of virtual machines in the recycling station to realize the reuse of abandoned virtual machine resources is specifically as follows:

[0015] Use the native soft-delete method of the OpenStack community to delete virtual machines. The recycling station saves the deleted virtual machines. In a platform using KVM / QEMU virtualization, the corresponding state of the deleted virtual machine is powered off. The virtual machines in the recycling station only occupy storage resources, and the CPU and memory physical resources are released;

[0016] The VIM component matches the virtual machines in the recycling station according to information such as network element type and version number. The interface used is GET / servers / detail?recycled=True&app_type=<network element type>&app_version=<network element version>. If available virtual machines are matched, try to reuse the virtual machines in the recycling station; since the virtual machines in the recycling station are in the powered-off state and can only be restored to run on their corresponding physical machines, it is necessary to first judge whether the corresponding CPU or memory physical resources of their physical machines are sufficient. If the physical resources are sufficient, restore the virtual machines in the recycling station through the VIM's virtual machine restoration interface POST / servers / {server_id} / action, and specify the body parameter {"restore":null}.

[0017] Furthermore, if no virtual machines meeting the requirements are matched in the recycling station, or if they are matched but the physical resources of their physical machines are insufficient, or if the expansion configuration requirements still cannot be met after restoring some virtual machines in the recycling station, then use the technology of introducing virtual machine cloning to complete the rapid expansion of network element virtual machines.

[0018] The introduction of virtual machine cloning technology to complete the rapid expansion of network element virtual machines is specifically as follows:

[0019] The VIM component passes through the virtual machine cloning interface POST / servers interface and specifies the body as {"clone":{"name":"<vm_new>","clone_type":"link_clone","max_count":"<number of virtual machines>",<other relevant cloning parameters specified according to the service>}} to complete the expansion of the existing network element virtual machines.

[0020] Furthermore, if the cloning of some virtual machines fails, the expansion of the network element capabilities is achieved by creating new virtual machines.

[0021] The introduction of virtual machine vertical dynamic scaling technology to achieve the rapid scaling of network element virtual machines includes vertical scaling down of virtual machine configurations and horizontal scaling down of virtual machine configurations, specifically as follows:

[0022] After the VIM component receives the network element virtual machine scaling down message, it will first determine whether the current virtual machine configuration is the lowest configuration. If it is not the lowest configuration, it will dynamically reduce the CPU or memory configuration of the virtual machine through the VIM virtual machine dynamic scaling interface. The relevant interface is the POST / servers / {server_id} / action interface, and the body is specified as {"live_resize":{"vcpus":"<CPU reduction value>","rams":"<RAM reduction value>"}}. According to the predetermined reduction value, the CPU or memory resources of the network element virtual machine are vertically reduced to complete the scaling down of the network element capabilities;

[0023] If it is already the lowest configuration, it will directly use the VIM virtual machine deletion interface DELETE / servers / {server_id} and specify the body as {"recycled":true} to delete the network element virtual machine to the recycle bin, thereby achieving the scaling down of the network element virtual machine.

[0024] The second aspect of this application provides a system for achieving rapid scaling of NFV network elements at the VI layer, including:

[0025] A vertical expansion module for introducing virtual machine vertical dynamic scaling technology to achieve rapid scaling of network element virtual machines;

[0026] A recycle bin reuse module for reusing virtual machines in the recycle bin to achieve the reuse of discarded virtual machine resources;

[0027] A cloned virtual machine module for introducing virtual machine cloning technology to complete the rapid expansion of network element virtual machines.

[0028] The third aspect of the present application provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for realizing rapid expansion and contraction of NFV network elements at the VI layer as described in the first aspect of the present application is implemented.

[0029] The fourth aspect of the present application provides a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the method for realizing rapid expansion and contraction of NFV network elements at the VI layer as described in the first aspect of the present application.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] Compared with the existing way of creating virtual machines, the present invention can realize the rapid expansion of NFV network elements at the VI layer by combining multiple strategies.

[0032] The present invention completes the rapid expansion and contraction of network element virtual machines by introducing the technology of vertical dynamic increase and decrease of virtual machines; reuses the virtual machines in the recycling station to realize the reuse of abandoned virtual machine resources; introduces the virtual machine cloning technology to complete the rapid expansion of network element virtual machines, replacing the virtual machine creation technology, thereby improving the expansion efficiency.

[0033] The solution of the present invention can reduce the amount of initialization configuration parameters for network element parameters; using the solution of the present invention, the time required for expansion and contraction of NFV network elements can be optimized from minutes to seconds. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a system module diagram of an implementation mode of the present invention.

[0035] Figure 2 This is ETSI's NFV reference architecture diagram.

[0036] Figure 3 It is a flow chart of expanding the capacity of a network element virtual machine in the present invention.

[0037] Figure 4 It is a flow chart of shrinking the network element virtual machine in the present invention. DETAILED DESCRIPTION

[0038] The above contents of the present invention are further described in detail below in the form of embodiments, but this should not be understood as the scope of the above subject matter of the present invention being limited to the following embodiments, and all technologies realized based on the above contents of the present invention belong to the scope of the present invention.

[0039] Some of the terms in the present invention are explained as follows:

[0040] MANO: Management and Orchestration

[0041] NFV: Network Functions Virtualization

[0042] NFVO: NFV Orchestrator, Network Function Virtualization Orchestration

[0043] VNF: Virtualised Network Function

[0044] VNFM: VNF Manager, Virtual Network Function Management

[0045] VI: Virtualized Infrastructure

[0046] VIM: Virtualized Infrastructure Manager

[0047] ETSI: European Telecommunications Standards Institute

[0048] The present invention provides a method for realizing rapid expansion and contraction of NFV network elements at the VI layer, comprising the following steps:

[0049] Introducing virtual machine vertical dynamic increase and decrease configuration technology to achieve rapid expansion and contraction of network element virtual machines;

[0050] Reuse virtual machines in the recycle bin to reuse abandoned virtual machine resources;

[0051] The virtual machine cloning technology is introduced to achieve rapid expansion of network element virtual machines.

[0052] The present invention is further described in detail below with reference to specific embodiments. Figure 1-4 As shown, the following solutions are included:

[0053] (I) Solutions for expanding NFV network elements

[0054] The monitoring and alarm module is interconnected with the VIM component through the Or-Vi interface. The Or-Vi interface supports RESTful, the message format is JSON, and the encoding is UTF-8. The monitoring and alarm module periodically obtains the real-time resource usage information of the network element virtual machine through the POST / v2 / monitor / realtimePerf interface, including: CPU usage (vm.cpu.usage) and memory usage (vm.memory.usage). At the same time, the monitoring and alarm module will receive the resource usage upper limit indicator issued by the resource orchestration module. After each periodic collection of the real-time usage information of the network element virtual machine resources, it will be compared with the usage upper limit. Once the CPU or memory usage exceeds the upper limit, an alarm is triggered, and the alarm information will be sent to the resource orchestration module through the message queue.

[0055] Among them, the VIM component is mainly used to manage resources in the virtualized infrastructure (NFVI). NFVI includes computing, storage and network resources. The monitoring module obtains monitoring data, such as CPU and memory usage data, through the API interface provided by VIM. At the same time, VIM also provides virtual machine management interfaces through RESTful API, such as virtual machine creation, deletion, recovery, expansion, reduction, cloning and other interfaces.

[0056] The resource orchestration module is mainly used for the overall orchestration management of the NFV network element virtual machine. It obtains the alarm message sent by the monitoring and alarm module through the message queue, and completes the monitoring and management of the network element virtual machine through the monitoring and alarm module and the resource management module. First, the orchestration module will receive the upper limit of the resource utilization rate issued by the OSS / BSS system, that is, the upper limit of the CPU utilization rate and the memory utilization rate (threshold_upper_cpu and threshold_upper_ram); secondly, the orchestration module will define the corresponding expansion configuration (including vertical expansion configuration and horizontal expansion configuration) according to different network element requirements, and finally, the orchestration module will send the upper limit of resource utilization rate to the monitoring and alarm module. After receiving the resource alarm notification, the orchestration module will send the alarm information of the expanded network element virtual machine and the corresponding expansion configuration of the network element to the resource management module through the message queue.

[0057] Among them, the resource management module, monitoring alarm module and resource management module are all designed and implemented in the MANO system. MANO (Management and Orchestration) is a management and orchestration system for NFV (Network Functions Virtualization), which is used to manage and orchestrate network function virtualization resources and services. Its main function is to achieve automated, elastic and flexible network service delivery, thereby improving the efficiency and service quality of network operators.

[0058] The capacity expansion configuration of the network element virtual machine includes vertical capacity expansion configuration and horizontal capacity expansion configuration. Vertical capacity expansion means not increasing the number of virtual machines, but only increasing the configuration of the virtual machine, such as increasing the number of CPU cores or increasing the memory value of the virtual machine; horizontal capacity expansion means not modifying the configuration of the virtual machine, but only increasing the number of virtual machines.

[0059] Let R0 be the resource utilization rate when the service is unloaded, R max is the resource usage upper limit (threshold_upper_cpu or threshold_upper_ram), r1 is the current resource usage ratio and R max The difference between R and R2 is the resource utilization rate after expansion. max The difference between N0 and N1 is the default configuration of the resource. c is the current configuration of the resource (number of CPU cores or memory size / G), n V+ is the expected vertical increase value, n H+ is the expected horizontal increase value (i.e. the number of newly added virtual machines), and G is the business load. The calculation method of the increase value of the two expansion methods is as follows:

[0060] (1) For the vertical expansion of virtual machines, the additional configuration value is calculated as follows:

[0061] ①G=N C ×(R max -R0+r1);

[0062] ②G=(N C +n V+ )×(R max -R0-r2)

[0063] Both ① and ② above use resource configuration × resource utilization rate to represent the business load, (N C +n V+ ) represents the configuration of resources after vertical addition. Combining the above ① and ②, we can get: From the results, we can see that the resource allocation value is proportional to the current configuration. The higher the resource overutilization rate and the lower the resource reduction degree, the greater the resource allocation value.

[0064] In actual operation, the minimum value for CPU resources is 1 core, and the minimum value for memory resources can be determined according to the business, and 1G is recommended. Therefore, the actual virtual machine configuration value is

[0065] (2) For the method of horizontally expanding the capacity of virtual machines, the additional configuration value is calculated as follows:

[0066] ①G=N C ×(Rmax -R0 + r1);

[0067] ②G = (N C + n H+ × N0) × (R max -R0 - r2)

[0068] Both ① and ② above represent the load of the service by the allocation of resources × the utilization rate of resources. (N C + n H + × N0) represents the allocation of resources after horizontal addition, that is, the total allocation of multiple virtual machines. Combining ① and ② above, it can be obtained that: As can be seen from the results, the value of resource addition is directly proportional to the current allocation and inversely proportional to the default allocation of resources. The higher the over-utilization rate of resources and the lower the reduction degree of resources, the greater the value of resource addition, that is, the more virtual machines need to be newly added.

[0069] In actual operation, for virtual machines, the minimum value of addition is 1 virtual machine. Therefore, the actual number of newly added virtual machines is

[0070] In some embodiments, the resource management module is interconnected with the VIM component through the Or-Vi interface, mainly for the life cycle management of network element virtual machines. After receiving the message of expanding the network element virtual machine and the corresponding expansion configuration of the network element, the resource management module will adopt various strategies to expand the network element virtual machine, so as to achieve the rapid expansion of the network element virtual machine. The relevant strategies are as follows:

[0071] 1. Vertically expand the virtual machine configuration

[0072] After receiving the message of expanding the network element virtual machine, the VIM component will first compare the CPU or memory resources of the physical machine where the network element virtual machine is located monitored with the configuration requirements of horizontal expansion.

[0073] If the physical resources are sufficient, through the VIM's POST / servers / {server_id} / action interface, and specify the body as {"live_resize": {"vcpus": "<CPU addition value>", "rams": "<RAM addition value>"}}, vertically add the CPU or memory resources of the network element virtual machine according to the predetermined addition value. This solution does not require shutting down the original virtual machine, creating a new virtual machine, has the fastest expansion speed of the network element virtual machine, does not pre-occupy any resources before expansion, and can complete the expansion of the network element's capabilities without initializing the configuration of network element parameters.

[0074] If the physical resources are insufficient, subsequent operations will be performed.

[0075] 2. Reuse the virtual machines in the recycle bin

[0076] The virtual machine recycle bin is not a function natively supported by the OpenStack community. The VIM functions in this patent have enhanced the native functions. The recycle bin is used to save deleted virtual machines. The virtual machines are deleted using the OpenStack community's native soft-delete method. In the platform using KVM / QEMU virtualization, the corresponding state of the deleted virtual machine is shutdown. Therefore, the virtual machines in the recycle bin will only occupy storage resources, while CPU and memory resources will be released.

[0077] The VIM component matches the virtual machine in the recycle bin according to the network element type, version number and other information. The interface used is GET / servers / detail? recycled=True&app_type=<network element type>&app_version=<network element version> (& and other filtering conditions). If an available virtual machine is matched, the virtual machine in the recycle bin is tried to be reused. However, since the virtual machine in the recycle bin is in a shutdown state, it can only be restored to the physical machine where it is located. Therefore, it is necessary to first determine whether the corresponding CPU or memory resources of the physical machine where it is located are sufficient. If the resources are sufficient, the virtual machine in the recycle bin can be restored through the VIM virtual machine recovery interface POST / servers / {server_id} / action and specify the body parameter {"restore":null}. This solution does not need to create a new virtual machine, but only needs to start the existing virtual machine in the recycle bin. The expansion of network element virtual machines is fast. The virtual machines in the recycle bin before expansion only occupy the corresponding storage resources but not CPU and memory resources. In addition, the expansion of network element capabilities can be completed by initializing and configuring a small number of specified parameters of the network element.

[0078] If no virtual machine that meets the requirements is found in the recycle bin, or if a virtual machine that meets the requirements is found but the physical machine where the virtual machine resides does not have sufficient resources, or if the requirements for capacity expansion cannot be met after restoring some virtual machines in the recycle bin, perform subsequent operations.

[0079] 3. Clone the virtual machine

[0080] Virtual machine cloning is not a function natively supported by the OpenStack community. The VIM functions in this patent have enhanced the native functions. Virtual machine cloning technology can quickly create a copy that is exactly the same as an existing virtual machine, including the operating system, applications, and data. Virtual machine cloning can be divided into two main types: (1) Full cloning: Full cloning creates a complete copy that is independent of the original virtual machine, including the virtual machine's configuration, operating system, applications, and data. This type of cloning occupies a large amount of storage space because it copies all data, but allows the creation of independent virtual machine instances without affecting the operation of the original virtual machine. (2) Linked cloning: Linked cloning is a more efficient cloning method that creates a snapshot of a virtual machine and shares the storage content of the original virtual machine. The cloned virtual machine and the original virtual machine share the same base image and only contain the differences between the virtual machines, reducing storage requirements.

[0081] The VIM component uses the virtual machine clone interface POST / servers interface and specifies the body as {"clone":{"name":"<vm_new> ","clone_type":"link_clone","max_count":"<number of virtual machines>",<specify other related cloning parameters according to the business>}}, to complete the expansion of the existing network element virtual machine, it is recommended to use linked cloning to clone the virtual machine, which is more efficient and faster. This solution uses the method of cloning virtual machines to add new virtual machines, and the expansion of network element virtual machines is fast. No resources are pre-occupied before expansion, and only a small number of network element parameters need to be initialized and configured to complete the expansion of network element capabilities.

[0082] If cloning of some virtual machines fails, perform subsequent operations.

[0083] 4. Create a new virtual machine

[0084] The VIM component creates an interface POST / servers / {server_id} / action through the virtual machine and specifies the body as {"server":{"name":"<vm_new> ","imageRef":"<virtual machine image ID>","flavorRef":"<virtual machine configuration>","max_count":"<number of virtual machines>",<other relevant cloning parameters specified according to the business>}}, complete the expansion of the existing network element virtual machine by creating a new virtual machine. This solution uses the new virtual machine method to add a new virtual machine. The expansion speed of the network element virtual machine is slow, and no resources are pre-occupied before expansion. All network element parameters need to be initialized and configured to complete the expansion of network element capabilities.

[0085] The following is a comparison of four ways to expand the capacity of network element virtual machines:

[0086] Serial number Way Speed ​​of creating virtual machines Pre-occupied resources Network element parameter initialization amount 1 Vertical expansion of virtual machines Fastest Do not pre-empt any resources No initialization required 2 Recycle Bin to restore virtual machine Faster Only reserve storage resources A small amount of parameter initialization 3 Cloning a VM Faster Do not pre-empt any resources A small amount of parameter initialization 4 Create a new virtual machine Slowest Do not pre-empt any resources Initialize all parameters

[0087] (II) Solution for reducing the capacity of NFV network elements

[0088] The monitoring and alarm module will receive the resource utilization lower limit indicator issued by the resource orchestration module. After periodically collecting the real-time utilization information of the network element virtual machine resources, it will be compared with the utilization lower limit. Once the CPU or memory utilization is lower than the lower limit, an alarm will be triggered and the alarm information will be sent to the resource orchestration module through the message queue.

[0089] The orchestration module will receive the lower limit of resource utilization sent by the OSS / BSS system, that is, the lower limit of CPU utilization and memory utilization (threshold_lower_cpu and threshold_lower_ram); secondly, the orchestration module will define the corresponding shrinking configuration (including vertical shrinking configuration and horizontal shrinking configuration) according to different network element requirements, and finally, the orchestration module will send the lower limit of resource utilization to the monitoring alarm module. After receiving the resource alarm notification, the orchestration module will send the alarm information of the expanded network element virtual machine and the corresponding expansion configuration of the network element to the resource management module through the message queue.

[0090] To calculate the virtual machine's reduced capacity configuration, let R0 be the resource utilization rate when the business is unloaded, and R min is the resource usage upper limit (threshold_lower_cpu or threshold_lower_ram), r3 is the current resource usage ratio and R min The difference between R and R4 is the resource utilization rate after expansion. min The difference between N0 and N1 is the default configuration of the resource. C is the current configuration of the resource (number of CPU cores or memory size / G), n V- is the expected vertical reduction value, n H- is the expected horizontal reduction value, and G is the business load. The calculation method of the increase value of the two expansion methods is as follows:

[0091] (1) For the vertical reduction of virtual machines, the reduction value is calculated as follows:

[0092] ①G=N C ×(R min -R0-r3);

[0093] ②G=(N C -n V- )×(R min -R0+r4)

[0094] Combining the above ① and ②, it can be obtained that: In actual operation, for CPU resources, the minimum value of reduction is 1 core, while for memory resources, the minimum value of reduction can be determined according to the service, and it is recommended to be 1G. Moreover, in order not to affect the services of the virtual machine, it is required that the configuration of the virtual machine after reduction cannot be lower than N0. Therefore, the actual virtual machine reduction value is

[0095] (2) For the method of horizontally scaling in virtual machines, since the loads of multiple virtual machines running the same service may be unbalanced and the impact of scaling in on the service is relatively large, a strategy of scaling in 1 virtual machine at a time is adopted to complete the scaling in.

[0096] In some embodiments, after the resource management module receives the message of scaling in the virtual machine of the network element and the scaling configuration corresponding to the network element, it will adopt multiple strategies to scale in the virtual machine of the network element. The relevant strategies are as follows:

[0097] 1. Vertical scaling in of virtual machine configuration

[0098] After the VIM component receives the message of scaling in the virtual machine of the network element, it will first determine whether the current configuration of the virtual machine is the lowest configuration. If it is not the lowest configuration, it will dynamically reduce the configuration of the CPU or memory of the virtual machine through the dynamic scaling interface of the virtual machine of VIM. The relevant interface is the POST / servers / {server_id} / action interface, and the body is specified as {"live_resize":{"vcpus":"<CPU reduction value>","rams":"<RAM reduction value>"}}, and vertically reduce the CPU or memory resources of the virtual machine of the network element according to the predetermined reduction value, so as to complete the scaling in of the network element capabilities.

[0099] 2. Horizontal scaling in of virtual machine configuration

[0100] If it is already the lowest configuration, the virtual machine of the network element will be directly deleted to the recycle bin through the virtual machine deletion interface DELETE / servers / {server_id} of VIM, and the body is specified as {"recycled":true}, so as to realize the scaling in of the virtual machine of the network element;

[0101] This application also provides a system for realizing rapid scaling of NFV network elements at the VI layer, including:

[0102] A vertical expansion module, which is used to introduce the technology of dynamically increasing and decreasing the vertical allocation of virtual machines to realize the rapid scaling of the virtual machine of the network element;

[0103] A recycle bin reuse module, which is used to reuse the virtual machines in the recycle bin to realize the reuse of the discarded virtual machine resources;

[0104] A cloned virtual machine module, which is used to introduce the virtual machine cloning technology to complete the rapid expansion of the virtual machine of the network element.

[0105] The present application also provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for realizing rapid expansion and contraction of NFV network elements at the VI layer as described above is implemented.

[0106] The present application also provides a computer-readable storage medium storing a computer program, wherein the computer program enables a computer to execute the method for realizing rapid expansion and contraction of NFV network elements at the VI layer as described above.

[0107] In the embodiments disclosed in the present application, the computer storage medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. The computer storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the above. More specific examples of computer storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above.

[0108] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any technician familiar with the profession, without departing from the scope of the technical solution of the present invention, according to the technical essence of the present invention, any simple modification, equivalent replacement and improvement made to the above embodiment still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for quickly scaling NFV network elements in the VI layer, comprising the following steps: Introduce the technology of dynamically increasing or decreasing the vertical configuration of virtual machines to achieve quick scaling of network element virtual machines; Reuse the virtual machines in the recycle bin to realize the reuse of discarded virtual machine resources; Introduce the virtual machine cloning technology to complete the quick expansion of network element virtual machines.

2. The method for realizing rapid expansion and contraction of NFV network elements at the VI layer according to claim 1, characterized in that: The introduction of the technology of dynamically increasing or decreasing the vertical configuration of virtual machines to achieve quick scaling of network element virtual machines includes vertical expansion of virtual machine configuration, specifically: Through the VIM's POST / servers / {server_id} / action interface, and specifying the body as {"live_resize":{"vcpus":"<CPU increment value>","rams":"<RAM increment value>"}}, vertically increment the physical resources of the network element virtual machine according to the predetermined increment value, where the physical resources include CPU and memory.

3. The method for realizing rapid expansion and contraction of NFV network elements at the VI layer according to claim 2, characterized in that: After the VIM component receives the network element virtual machine expansion message, compare the monitored CPU or memory physical resources of the physical machine where the network element virtual machine is located with the horizontal expansion configuration requirements; If the physical resources are sufficient, use the technology of dynamically increasing or decreasing the vertical configuration of virtual machines to achieve quick scaling of network element virtual machines; if the physical resources are insufficient, reuse the virtual machines in the recycle bin to realize the reuse of discarded virtual machine resources.

4. The method for realizing rapid expansion and contraction of NFV network elements at the VI layer according to claim 1, characterized in that: The reuse of the virtual machines in the recycle bin to realize the reuse of discarded virtual machine resources is specifically: Use the native soft-delete method of the OpenStack community to delete virtual machines. The recycle bin saves the deleted virtual machines. In a platform using KVM / QEMU virtualization, the corresponding state of the deleted virtual machine is shutdown. The virtual machines in the recycle bin only occupy storage resources, and the CPU and memory physical resources are released; The VIM component matches the virtual machines in the recycle bin according to information such as network element type and version number. The interface used is GET / servers / detail?recycled=True&app_type=<network element type>&app_version=<network element version>. If available virtual machines are matched, try to reuse the virtual machines in the recycle bin; since the virtual machines in the recycle bin are in the shutdown state, they can only be restored to run on their corresponding physical machines. Therefore, it is necessary to first determine whether the corresponding CPU or memory physical resources of their physical machines are sufficient. If the physical resources are sufficient, use the VIM's virtual machine restoration interface POST / servers / {server_id} / action and specify the body parameter {"restore":null} to restore the virtual machines in the recycle bin.

5. The method for realizing rapid expansion and contraction of NFV network elements at the VI layer according to claim 4, characterized in that: If no virtual machines meeting the requirements are matched in the recycle bin, or the physical resources of their corresponding physical machines are insufficient after matching, or after restoring some of the virtual machines in the recycle bin, the requirements for expansion configuration still cannot be met, then use the virtual machine cloning technology to complete the quick expansion of network element virtual machines.

6. The method for realizing rapid expansion and contraction of NFV network elements at the VI layer according to claim 1, characterized in that: The introduction of the virtual machine cloning technology to complete the quick expansion of network element virtual machines is specifically: The VIM component completes the expansion of the existing network element virtual machine through the virtual machine cloning interface POST / servers interface, and specifies the body as {"clone":{"name":"<vm_new>","clone_type":"link_clone","max_count":"<number of virtual machines>",<other relevant cloning parameters specified according to the business>}}.

7. The method for realizing rapid expansion and contraction of NFV network elements at the VI layer according to claim 1, characterized in that: The introduction of the virtual machine vertical dynamic scaling technology realizes the rapid scaling of the network element virtual machine, including vertical scaling down of the virtual machine configuration and horizontal scaling down of the virtual machine configuration, specifically: After the VIM component receives the network element virtual machine scaling down message, it will first determine whether the current virtual machine configuration is the lowest configuration. If it is not the lowest configuration, it will dynamically reduce the CPU or memory configuration of the virtual machine through the VIM virtual machine dynamic scaling interface. The relevant interface is the POST / servers / {server_id} / action interface, and the body is specified as {"live_resize":{"vcpus":"<CPU reduction value>","rams":"<RAM reduction value>"}}. According to the predetermined reduction value, vertically reduce the CPU or memory resources of the network element virtual machine, thereby completing the scaling down of the network element capabilities; If it is already the lowest configuration, directly use the VIM virtual machine deletion interface DELETE / servers / {server_id}, and specify the body as {"recycled":true}, and delete the network element virtual machine to the recycle bin, thereby realizing the scaling down of the network element virtual machine.

8. A system for realizing rapid expansion and contraction of NFV network elements at the VI layer, characterized in that: Including: A vertical expansion module for introducing the virtual machine vertical dynamic scaling technology to realize the rapid scaling of the network element virtual machine; A recycle bin reuse module for reusing the virtual machines in the recycle bin to realize the reuse of the discarded virtual machine resources; A cloned virtual machine module for introducing the virtual machine cloning technology to complete the rapid expansion of the network element virtual machine.

9. An electronic device, characterized in that: Including: A memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, it realizes the method for rapidly scaling the NFV network element at the VI layer as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, and the computer program causes a computer to execute the method for rapidly scaling the NFV network element at the VI layer as described in any one of claims 1-7.

Images