Virtual network function resource configuration method and device
Through the strategy generation unit based on machine learning algorithm, matching the appropriate resource pool, the problem of management information model difference between MEPM and VIM is solved, the adaptability and flexibility of the VNF resource pool are improved, and the precise allocation of resources is achieved. Efficient use.
Patent Information
- Application Number
- CN201911354213.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-12-25
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2039-12-25
AI Technical Summary
In the existing MEC system, the management information model and interface differences between MEPM and VIM cannot be effectively parsed by the management information required for edge computing management, and the resource pool type required for new VNF is not able to be evaluated, which limits the self-contained resource pool. Adaptation and flexibility.
Adopt a strategy generation unit based on machine learning algorithms, and obtain the resource characteristics and resource pool characteristics of VNF, match the appropriate target resource pool, realize the management interface adaptation of multi-vendor MEPM platform and multi-vendor VIM, and improve the adaptability of resource pools. degree and flexibility.
It realizes the adaptation of the multi-vendor MEPM platform and the multi-vendor VIM management interface, improves the adaptability and flexibility of the VNF service allocation resource pool, and ensures the precise allocation and efficient use of resources.
Smart Images

Figure CN113110884B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of NFV (Network Functions Virtualization), and in particular to a VNF (Virtualised Network Function) resource configuration method and device. Background Art
[0002] For the MEC (Mobile Edge Computing) system, the key point of the existing ETSI (European Telecommunications Standards Institute) solution is that when MEPM (Mobile Edge Platform Manager) and VIM (Virtualised Infrastructure Manager) communicate, matching vendor solutions must be deployed to avoid differences in management information models and interfaces. The resource pool type required for each VNF is formulated through a pre-set mode between MEPM and VIM. Summary of the invention
[0003] The inventors found that the solutions in the related art cannot evaluate the resource pool types required by the new VNF to guide the construction of resource pools that adapt to various types of VNFs, and the current MEC standard framework has put forward the requirement of unified management for VIM. There are large differences in the management interface and data model between the multi-vendor MEPM platform and the multi-vendor VIM. These differences make VIM unable to parse the management information required for edge computing management.
[0004] One purpose of the present disclosure is to improve the adaptability and flexibility of allocating resource pools for VNF services.
[0005] According to one aspect of some embodiments of the present disclosure, a VNF resource configuration method is proposed, including: obtaining resource characteristics required by the VNF; obtaining characteristics of each resource pool; selecting a target resource pool for the VNF according to the resource characteristics required by the VNF and the characteristics of the resource pool through a policy generation unit, wherein the policy generation unit is based on a machine learning algorithm, and is generated based on VIM resource management information of multiple manufacturers, and MEPM of multiple manufacturers' instantiation information and management information about VNF; and configuring a target resource pool for the VNF.
[0006] In some embodiments, configuring a target resource pool for a VNF includes: sending first configuration information to a MEPM corresponding to the VNF according to the target resource pool; and sending second configuration information to a VIM corresponding to the target resource pool.
[0007] In some embodiments, the machine learning algorithm includes at least one of a reinforcement learning algorithm Q-learning or a deep learning algorithm.
[0008] In some embodiments, the VNF resource configuration method also includes: obtaining VIM resource management information from multiple manufacturers; obtaining MEPM instantiation information and lifecycle management information about VNF from multiple manufacturers; and generating a policy generation unit based on machine learning algorithm training according to the VIM resource management information and the MEPM instantiation information and management information about VNF from multiple manufacturers.
[0009] Through this method, the characteristics of the resources required by the VNF and the characteristics of each resource pool can be determined, and the resource pool suitable for the VNF can be extracted based on machine learning algorithm matching, so as to achieve management interface adaptation between multi-vendor MEPM platforms and multi-vendor VIMs, and improve the adaptability and flexibility of resource pool allocation for VNF services.
[0010] According to one aspect of some other embodiments of the present disclosure, a VNF resource configuration device is proposed, including: a required resource feature acquisition unit, configured to acquire resource features required by the VNF; a resource pool feature acquisition unit, configured to acquire features of each resource pool; a policy generation unit, configured to select a target resource pool for the VNF based on the resource features required by the VNF and the features of the resource pool, wherein the policy generation unit is based on a machine learning algorithm, and is generated based on VIM resource management information of multiple manufacturers, and MEPM of multiple manufacturers' instantiation information and management information about VNFs; and a configuration unit, configured to configure the target resource pool for the VNF.
[0011] In some embodiments, the configuration unit is configured to: send first configuration information to the MEPM corresponding to the VNF according to the target resource pool; and send second configuration information to the VIM corresponding to the target resource pool.
[0012] In some embodiments, the machine learning algorithm includes at least one of a reinforcement learning algorithm Q-learning or a deep learning algorithm.
[0013] In some embodiments, the VNF resource configuration device also includes: a training unit, configured to: obtain VIM resource management information from multiple manufacturers; obtain MEPM instantiation information and lifecycle management information about VNF from multiple manufacturers; and generate a policy generation unit based on machine learning algorithm training according to the VIM resource management information and MEPM instantiation information and management information about VNF from multiple manufacturers.
[0014] According to one aspect of some other embodiments of the present disclosure, a VNF resource configuration device is proposed, including: a memory; and a processor coupled to the memory, the processor being configured to execute any one of the above VNF resource configuration methods based on instructions stored in the memory.
[0015] Such a VNF resource configuration device can determine the characteristics of the resources required by the VNF and the characteristics of each resource pool, and extract the resource pool suitable for the VNF based on machine learning algorithm matching, thereby realizing the management interface adaptation of multi-vendor MEPM platforms and multi-vendor VIMs, and improving the adaptability and flexibility of allocating resource pools for VNF services.
[0016] According to an aspect of some further embodiments of the present disclosure, a computer-readable storage medium is proposed, on which computer program instructions are stored, and when the instructions are executed by a processor, the steps of any one of the VNF resource configuration methods described above are implemented.
[0017] By executing instructions on such a computer-readable storage medium, the characteristics of the resources required by the VNF and the characteristics of each resource pool can be determined, and the resource pool suitable for the VNF can be extracted based on machine learning algorithm matching, thereby achieving management interface adaptation between multi-vendor MEPM platforms and multi-vendor VIMs, and improving the adaptability and flexibility of allocating resource pools for VNF services.
[0018] In addition, according to one aspect of some embodiments of the present disclosure, a MEC system is proposed, comprising a plurality of MEPMs; a plurality of VIMs; and any one of the VNF resource configuration devices mentioned above.
[0019] Such an MEC system can determine the characteristics of the resources required by the VNF and the characteristics of each resource pool, and extract the resource pool suitable for the VNF based on machine learning algorithm matching, thereby enabling the management interface adaptation of multi-vendor MEPM platforms and multi-vendor VIMs, and improving the adaptability and flexibility of allocating resource pools for VNF services. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the drawings:
[0021] Figure 1 Flowcharts of some embodiments of the VNF resource configuration method disclosed in the present invention.
[0022] Figure 2 Flowcharts of some embodiments of the training process in the VNF resource configuration method disclosed in the present invention.
[0023] Figure 3 Schematic diagrams of some embodiments of the VNF resource configuration device disclosed in the present invention.
[0024] Figure 4 Schematic diagrams of other embodiments of the VNF resource configuration device disclosed in the present invention.
[0025] Figure 5 Schematic diagrams of some further embodiments of the VNF resource configuration device disclosed in the present invention.
[0026] Figure 6 Schematic diagrams of some embodiments of the MEC system disclosed herein. DETAILED DESCRIPTION
[0027] The technical solution of the present disclosure is further described in detail below through the accompanying drawings and embodiments.
[0028] The flowcharts of some embodiments of the VNF resource configuration method disclosed in the present invention are as follows Figure 1 shown.
[0029] In step 101, the resource characteristics required by the VNF are obtained. In some embodiments, the service characteristics of the VNF can be obtained, and the characteristics of the required resources are determined based on the service characteristics. In some embodiments, the service characteristics of the VNF can be input into the policy generation unit as the required resource characteristics.
[0030] In step 102, the characteristics of each resource pool are obtained. In some embodiments, the characteristics of the resource pool may include vendor, interface, capacity, computing resources, storage resources, network resources, hardware acceleration resources, etc.
[0031] In step 103, a target resource pool is selected for the VNF by a policy generation unit according to the resource characteristics required by the VNF and the characteristics of the resource pool. In some embodiments, the policy generation unit is based on a machine learning algorithm, and is trained and generated according to VIM resource management information of multiple manufacturers, and MEPM instantiation information and management information about VNF of multiple manufacturers.
[0032] In step 104, a target resource pool is configured for the VNF. In some embodiments, according to the target resource pool, first configuration information is sent to the MEPM corresponding to the VNF; second configuration information is sent to the VIM corresponding to the target resource pool, thereby sending configuration information to both ends of the connection respectively, so that the MEPM and the VIM corresponding to the target resource pool can communicate with each other.
[0033] Through this method, the characteristics of the resources required by the VNF and the characteristics of each resource pool can be determined, and the resource pool suitable for the VNF can be extracted based on machine learning algorithm matching, so as to achieve management interface adaptation between multi-vendor MEPM platforms and multi-vendor VIMs, and improve the adaptability and flexibility of resource pool allocation for VNF services.
[0034] The flowcharts of some embodiments of the training process in the VNF resource configuration method disclosed in the present invention are as follows Figure 2 shown.
[0035] In step 201, VIM resource management information of multiple manufacturers is obtained. In some embodiments, the obtained VIM resource management information can be translated and modeled to facilitate analysis and management.
[0036] In step 202, the instantiation information and lifecycle management information of the VNFs of MEPMs of multiple vendors are obtained. In some embodiments, the instantiation information and lifecycle management information obtained can be modeled to facilitate analysis and management.
[0037] In step 203, a policy generation unit is generated based on the VIM resource management information and the instantiation information and management information of the VNFs of the MEPMs of multiple vendors and based on the machine learning algorithm training. In some embodiments, the machine learning algorithm may include at least one of a reinforcement learning Q-learning algorithm and a deep learning algorithm.
[0038] When it is necessary to scale the capacity of a VNF or allocate a new resource pool to a VNF, the required virtual deployment resources are feature extracted and marked, and multiple VNF instantiation and scaling templates are extracted and marked. Based on the resource features required by the VNF, the features of the resources included in the system, and the policy generation unit based on algorithms such as Q-learning, a matching degree is calculated to obtain a target resource pool that matches the VNF, thereby providing deployment recommendations such as initialization deployment, scaling, etc. for the VNF, achieving accurate resource allocation and significantly improving overall resource utilization.
[0039] Schematic diagrams of some embodiments of the VNF resource configuration device disclosed in the present invention are as follows Figure 3 shown.
[0040] The required resource feature acquisition unit 301 can acquire the required resource features of the VNF. In some embodiments, the service features of the VNF can be acquired, and the features of the required resources can be determined based on the service features. In some embodiments, the service features of the VNF can be used as the required resource features.
[0041] The resource pool feature acquisition unit 302 can acquire the features of each resource pool. In some embodiments, the features of the resource pool may include vendor, interface, capacity, computing resources, storage resources, network resources, hardware acceleration resources, etc.
[0042] The policy generation unit 303 can select a target resource pool for the VNF according to the resource characteristics required by the VNF and the characteristics of the resource pool through the policy generation unit. In some embodiments, the policy generation unit is based on a machine learning algorithm, based on VIM resource management information of multiple manufacturers, and MEPM of multiple manufacturers about the instantiation information and management information of the VNF.
[0043] The configuration unit 304 can configure a target resource pool for the VNF. In some embodiments, according to the target resource pool, first configuration information is sent to the MEPM corresponding to the VNF; second configuration information is sent to the VIM corresponding to the target resource pool, thereby sending configuration information to both ends of the connection respectively, so that the MEPM and the VIM corresponding to the target resource pool can communicate with each other.
[0044] Such a VNF resource configuration device can determine the characteristics of the resources required by the VNF and the characteristics of each resource pool, and extract the resource pool suitable for the VNF based on machine learning algorithm matching, thereby realizing the management interface adaptation of multi-vendor MEPM platforms and multi-vendor VIMs, and improving the adaptability and flexibility of allocating resource pools for VNF services.
[0045] In some embodiments, the VNF resource configuration device may further include a training unit 305, which can generate a strategy generation unit based on machine learning algorithm training according to VIM resource management information and instantiation information and management information of VNFs of MEPMs of multiple manufacturers. In some embodiments, the machine learning algorithm may include at least one of a reinforcement learning Q-learning algorithm and a deep learning algorithm.
[0046] Such a VNF resource configuration device can calculate the matching degree according to the resource characteristics required by the VNF, the characteristics of the resources included in the system, and the policy generation unit based on algorithms such as Q-learning, and obtain a target resource pool that matches the VNF, thereby providing deployment recommendations such as initialization deployment, expansion and contraction of the VNF, achieving accurate allocation of resources, and significantly improving overall resource utilization.
[0047] A structural diagram of an embodiment of the VNF resource configuration device disclosed in the present invention is as follows Figure 4 As shown. The VNF resource configuration device includes a memory 401 and a processor 402. Among them: the memory 401 can be a disk, a flash memory or any other non-volatile storage medium. The memory is used to store the instructions in the corresponding embodiment of the VNF resource configuration method above. The processor 402 is coupled to the memory 401 and can be implemented as one or more integrated circuits, such as a microprocessor or a microcontroller. The processor 402 is used to execute instructions stored in the memory, which can improve the adaptability and flexibility of allocating resource pools for VNF services.
[0048] In one embodiment, it is also possible to Figure 5 As shown, the VNF resource configuration device 500 includes a memory 501 and a processor 502. The processor 502 is coupled to the memory 501 via a BUS bus 503. The VNF resource configuration device 500 can also be connected to an external storage device 505 via a storage interface 504 to call external data, and can also be connected to a network or another computer system (not shown) via a network interface 506. No further detailed introduction is given here.
[0049] In this embodiment, by storing data instructions in a memory and then processing the instructions through a processor, the adaptability and flexibility of allocating resource pools for VNF services can be improved.
[0050] In another embodiment, a computer-readable storage medium stores computer program instructions, which, when executed by a processor, implement the steps of the method in the corresponding embodiment of the VNF resource configuration method. Those skilled in the art should understand that the embodiments of the present disclosure may be provided as methods, devices, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable non-transitory storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0051] Schematic diagrams of some embodiments of a portion of the MEC system of the present disclosure are as follows Figure 6 As shown, multiple MEPMs 611-61n, n is a positive integer. MEPMs 611-61n may come from different manufacturers and have different interface parameters; multiple VIMs 621-62m, m is a positive integer, VIMs 621-62m may come from different manufacturers and have different interface parameters.
[0052] The VNF resource configuration device 601 is located between the MEPM 611~61n and the VIM 621~62m, and can use the VNF lifecycle management information collector and the virtual resource management information collector to obtain VFN and virtual resource management information respectively; extract VNF and resource features through template feature marking; use the policy generation unit based on Q-learning or deep learning algorithm to infer and calculate the resource type that needs to be deployed under the VNF, and recommend the target resource pool.
[0053] Such an MEC system can determine the characteristics of the resources required by the VNF and the characteristics of each resource pool, and extract the resource pool suitable for the VNF based on machine learning algorithm matching, thereby enabling the management interface adaptation of multi-vendor MEPM platforms and multi-vendor VIMs, and improving the adaptability and flexibility of allocating resource pools for VNF services.
[0054] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0055] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0056] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0057] So far, the present disclosure has been described in detail. In order to avoid obscuring the concept of the present disclosure, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solution disclosed here.
[0058] The method and apparatus of the present disclosure may be implemented in many ways. For example, the method and apparatus of the present disclosure may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above order of steps for the method is for illustration only, and the steps of the method of the present disclosure are not limited to the order specifically described above, unless otherwise specifically stated. In addition, in some embodiments, the present disclosure may also be implemented as a program recorded in a recording medium, which includes machine-readable instructions for implementing the method according to the present disclosure. Therefore, the present disclosure also covers a recording medium storing a program for executing the method according to the present disclosure.
[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present disclosure rather than to limit it. Although the present disclosure has been described in detail with reference to the preferred embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present disclosure can still be modified or some technical features can be replaced by equivalents without departing from the spirit of the technical solution of the present disclosure, which should be included in the scope of the technical solution for protection of the present disclosure.
Claims
1. A virtual network function resource configuration method, include: Obtain resource characteristics required by virtual network function VNF; Get the characteristics of each resource pool; A target resource pool is selected for the VNF by a policy generation unit according to the resource characteristics required by the VNF and the characteristics of the resource pool, wherein the policy generation unit is generated based on a machine learning algorithm and trained according to VIM resource management information of multiple vendors and MEPM instantiation information and management information about VNF of multiple vendors; Configure the target resource pool for the VNF.
2. The method according to claim 1, in, Configuring the target resource pool for the VNF includes: According to the target resource pool, send first configuration information to the MEPM corresponding to the VNF; Send the second configuration information to the VIM corresponding to the target resource pool.
3. The method according to claim 1, in, The machine learning algorithm includes at least one of a reinforcement learning algorithm Q-learning or a deep learning algorithm.
4. The method according to claim 1, further comprising: include: Obtain VIM resource management information from multiple vendors; Obtain VNF instantiation and lifecycle management information from MEPMs of multiple vendors; According to the VIM resource management information, as well as the instantiation information and management information of VNFs from MEPMs of multiple vendors, the policy generation unit is generated based on the machine learning algorithm training.
5. A virtual network function resource configuration device, include: A required resource feature acquisition unit is configured to acquire resource features required by a virtual network function VNF; A resource pool characteristic acquisition unit, configured to acquire characteristics of each resource pool; A policy generating unit is configured to select a target resource pool for the VNF according to the resource characteristics required by the VNF and the characteristics of the resource pool, wherein the policy generating unit is generated based on a machine learning algorithm and trained according to VIM resource management information of multiple vendors and MEPM instantiation information and management information about VNF of multiple vendors; A configuration unit is configured to configure the target resource pool for the VNF.
6. The device according to claim 5, in, The configuration unit is configured to: According to the target resource pool, send first configuration information to the MEPM corresponding to the VNF; Send the second configuration information to the VIM corresponding to the target resource pool.
7. The device according to claim 5, in, The machine learning algorithm includes at least one of a reinforcement learning algorithm Q-learning or a deep learning algorithm.
8. The device according to claim 5, further comprising: include: The training unit is configured as: Obtain VIM resource management information from multiple vendors; Obtain VNF instantiation and lifecycle management information from MEPMs of multiple vendors; According to the VIM resource management information, as well as the instantiation information and management information of VNFs from MEPMs of multiple vendors, the policy generation unit is generated based on the machine learning algorithm training.
9. A virtual network function resource configuration device, include: Memory; as well as A processor coupled to the memory, the processor being configured to execute the method according to any one of claims 1 to 4 based on instructions stored in the memory.
10. A computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the steps of the method according to any one of claims 1 to 4.
11. A mobile edge computing system, include: Multiple multi-access edge cloud platforms manage MEPM; Multiple virtualized infrastructure managers (VIMs); and A virtual network function VNF resource configuration device as described in any one of claims 5 to 9.
Citation Information
Patent Citations
Method and device for configuring virtual resource
CN105955824A
Configuration method and device
CN109818867A
