Resource orchestration method and device for multi-access edge computing network
By receiving virtual network requests in a multi-access edge computing network and calculating virtual network mapping solutions based on resource requirements and current situations, the problem of virtual network mapping in the prior art is solved, and efficient resource utilization is achieved Effective mapping with virtual networks.
Patent Information
- Application Number
- CN202010718441.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Priority Date
- 2019-11-01
- Filing Date
- 2020-07-23
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2040-07-23
AI Technical Summary
The existing virtual network mapping (VNE) problems are mainly concentrated in data center environments, and it is difficult to effectively execute in multi-access edge computing network environments, and it is impossible to optimize the utilization of physical network resources.
Provide a resource orchestration method and device, by receiving virtual network requests, judging periodic resource orchestration time, calculate the virtual network mapping solution based on resource requirements, usage status, node attributes and link attributes, and handing it over to multiple accesses Control nodes of edge computing networks, or reject requests.
Effectively solve the virtual network mapping requirements in a multi-access edge computing network environment, improve resource utilization and processing efficiency, and meet the virtual network mapping requirements in distributed edge areas.
Smart Images

Figure CN112783643B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of computer communications, and in particular to a resource orchestration method and device for a multi-access edge computing network. Background Art
[0002] How to map multiple virtual networks to a physical network to optimize the utilization of physical network resources has always been a major challenge in network virtualization. This problem is usually called the Virtual Networking Embedding (VNE) problem.
[0003] Most VNE problems are built in data center environments or focus on offline solutions. However, such VNE problems and their solution algorithms cannot be executed in a multi-access edge computing network environment composed of multiple distributed edge servers. Summary of the invention
[0004] In view of this, it is necessary to provide a resource orchestration method and device for a multi-access edge computing network, which can solve a virtual network mapping solution in a multi-access edge computing network environment.
[0005] A resource orchestration method for a multi-access edge computing network, the method comprising: receiving a virtual network request; determining whether a time for periodic resource orchestration has arrived; when the time for periodic resource orchestration has arrived, calculating a first virtual network mapping solution for the virtual network request based on resource requirements of the virtual network request, resource usage status of the multi-access edge computing network, attributes of physical nodes, and attributes of physical links; determining whether the first virtual network mapping solution exists; and when determining that the first virtual network mapping solution exists, handing over the first virtual network mapping solution to at least one control node of the multi-access edge computing network, otherwise, rejecting the virtual network request.
[0006] A resource orchestration device for a multi-access edge computing network, characterized in that the device includes: a processing unit, and a computer-readable storage medium for storing at least one computer program, wherein the computer program includes instructions executed by the processing unit, so that the processing unit performs the following steps: receiving a virtual network request; determining whether the time for periodic resource orchestration has arrived; when the time for periodic resource orchestration has arrived, calculating and solving a first virtual network mapping solution for the virtual network request based on the resource requirements of the virtual network request, the current resource usage status of the multi-access edge computing network, the attributes of the physical nodes, and the attributes of the physical links; determining whether the first virtual network mapping solution exists; and when it is determined that the first virtual network mapping solution exists, handing over the first virtual network mapping solution to at least one control node of the multi-access edge computing network, otherwise, rejecting the virtual network request.
[0007] The resource orchestration device of the multi-access edge computing network in the present invention, after receiving a virtual network request, calculates and solves a first virtual network mapping solution for the virtual network request according to the resource requirements of the virtual network request, the current resource usage status of the multi-access edge computing network, the attributes of the entity nodes, and the attributes of the entity links, and hands over the solution to at least one control node of the multi-access edge computing network for implementation, thereby solving the virtual network mapping requirements of the multi-access edge computing network environment in multiple distributed edge areas. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] Figure 1 This is an architecture diagram of a multi-access edge computing network in one embodiment of the present invention.
[0009] Figure 2 Schematic diagram of message exchange in one embodiment of the present invention.
[0010] Figure 3A , 3B 3C, 3D, 3E, 3F, 3G, 3H and 3I are schematic diagrams of virtual network mapping in one embodiment of the present invention.
[0011] Figure 4 The figure is a flow chart of a resource arrangement method in one embodiment of the present invention.
[0012] Figure 5 The figure is a flow chart of a resource scheduling method in another embodiment of the present invention.
[0013] Figure 6 FIG. 4 is a block diagram of a resource scheduling device according to an embodiment of the present invention.
[0014] Main component symbols
[0015] Multi-access edge computing network 100 Orchestrator 102 Control Node 104A, 104B, 104C Compute Node A1~A6、B1~B7、C1~C6 Edge Area 110A, 110B, 110C Service Provider 200 Physical Network 300 Virtual network request 310 Virtual Node a, b, c Entity Node A, B, C step S402~S412、S502~S510 Resource orchestration device 600 Processing Unit 610 Computer readable storage medium 620
[0016] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION
[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.
[0019] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.
[0020] Please refer to Figure 1 , which is an architecture diagram of a multi-access edge computing network 100 in one embodiment of the present invention. The multi-access edge computing network 100 includes an orchestrator 102, a plurality of control nodes 104A, 104B, and 104C, a plurality of computing nodes A1-A6, B1-B7, and C1-C6, and a plurality of edge areas 110A, 110B, and 110C.
[0021] The orchestrator 102 is used to provide real-time resource orchestration, including configuration of virtual and physical network functions. The orchestrator 102 includes an access point that allows users and third parties to access to subscribe to applications or deploy servers. When the orchestrator 102 receives a virtual network request (Virtual Network Request, referred to as VNR), the orchestrator 102 calculates and solves a virtual network mapping (Virtual Network Embedding, referred to as VNE) solution based on resource usage information, and then hands over the VNE solution to at least one control node in 104A~104C to execute the VNE solution.
[0022] Each control node 104A, 104B and 104C is used to prepare a virtualized network and report resource usage information. Each control node 104A, 104B and 104C is configured to manage the resources of computing nodes A1-A6, B1-B7 and C1-C6 and the physical link resources in its edge areas 110A, 110B and 110C, such as resource allocation and power management. Each control node 104A, 104B and 104C is configured to manage computing nodes A1-A6, B1-B7 and C1-C6 in a specific geographical location. Due to management efficiency and security issues, control nodes are usually not allowed to access computing nodes in different edge areas. After receiving the VNE solution from the orchestrator 102, the control nodes 104A-104C will execute VNE for the managed edge areas 110A-110C. In this embodiment, the control nodes 104A-104C may be any type of virtualized infrastructure manager (VIM), such as OpenStack, Kubernetes, or ONAP, etc., which is used to control and manage infrastructure supporting virtualization with the permission of the operator of the infrastructure provider.
[0023] Each edge region 110A, 110B, and 110C may be communicatively connected to each other via physical links, but the propagation delay may be much higher than the propagation delay within the same edge region. Each edge region is managed by a control node, which performs VNE resource allocation in the edge region it manages (e.g., the edge region 110A and the control node 104A).
[0024] In one embodiment, when the orchestrator 102 receives a VNR from an edge area, for example, the edge area 110A, the orchestrator 102 calculates a VNE solution and then hands over the VNE solution to the control node closest to the source of the VNR. For example, the edge area 110A corresponds to the control node 104A, and the control node 104A can execute the VNE solution in the edge area 110A under its control and management. In one embodiment, if the VNE solution of the orchestrator 102 involves multiple edge areas, the orchestrator 102 can hand over the VNE solution to multiple corresponding control nodes for execution.
[0025] Please refer to Figure 2, which is a schematic diagram of the message exchange sequence between the service provider 200, the orchestrator 102, the control node 104A and the computing node A1 in one embodiment of the present invention. The orchestrator 102 receives VNR input from the service provider 200, and calculates and solves the VNE solution based on the processing requirements and resource usage information of the VNR. If there is no suitable VNE solution, the orchestrator 102 will reject the VNR. If there is a suitable VNE solution, the orchestrator 102 will hand over the VNE solution to one or more control nodes, such as the control node 104A. The control node 104A executes the VNE solution. After processing the VNR, all computing nodes managed by the control node 104A report the resource usage information to the control node 104A, and then forward the resource usage information to the orchestrator 102 via the control node 104A, so that the orchestrator 102 can update the resource usage status according to the received resource usage information and continue to solve new VNE solutions.
[0026] In one embodiment, the computing nodes and networks in the edge region represent physical nodes and physical networks, respectively. Each physical network is composed of physical nodes and physical links. Each physical node includes node attributes, and the node attributes include processing power and geographical area. In one embodiment, the processing power is the number of instructions that can be completed per unit time. The geographical area includes information about the edge region where the physical node is located. In one embodiment, each physical node also includes multiple cores, and each physical node can be configured as one of a standard mode and an asymmetric mode. When the physical node is configured as a standard mode, the multiple cores of the physical node all operate at the same standard working speed; when the physical node is configured as an asymmetric mode, the multiple cores of the physical node can operate at different working speeds. In one embodiment, when the physical node is configured as an asymmetric mode, some of the multiple cores can operate at a first working speed higher than the standard working speed, and the remaining cores can operate at a second working speed lower than the standard working speed. In one embodiment, when the physical node is configured as an asymmetric mode, the physical node can dynamically reduce the processing speed of low-priority tasks, thereby increasing the processing speed of high-priority tasks.
[0027] In one embodiment, the entity link between the entity nodes includes link attributes, wherein the link attributes include processing capability and latency, and the processing capability includes bandwidth.
[0028] In one embodiment, the virtual network is modeled as an Activity on Vertex topology, and the MEC application is modeled as a Service Chain topology. In one embodiment, a specific MEC application can be abstracted into multiple processes, and the traffic of the MEC application passes through the multiple processes in a predefined order.
[0029] In one embodiment, the VNR for the MEC application can be formatted as an AOV network request format, where the AOV network is used to represent the order of service of virtual nodes, vertices represent virtual nodes, and edges represent the order of service between virtual nodes. Any virtual node can only start processing the instructions of the virtual node after the virtual nodes and virtual links in its previous order have completed the processing process. When the virtual node and / or virtual link completes its process, the resources are released.
[0030] In one embodiment, the resource requirements of the VNR include priority, latency requirements, resource requirements of the virtual node, and resource requirements of the virtual link.
[0031] In one embodiment, one or more constraints may be set in the process of mapping a virtual network to a given physical network. In one embodiment, the constraints include a first constraint: VNRs with high priority or low latency requirements (e.g., video applications or vehicle-to-vehicle communications) can only be mapped to high-speed or standard-speed cores to meet the quality of user experience (Quality of Experience, referred to as QoE), and a second constraint: VNRs with low priority or insensitive to latency (e.g., specific IoT applications) can only be mapped to standard-speed or low-speed cores when the first virtual node is mapped for the first time. The control node monitors the computing resources in the edge area it controls and manages, and reports back to the orchestrator. Once the high-speed core of the physical node has available computing resources, the orchestrator can switch the virtual nodes that have been mapped but have not yet completed the process to the high-speed core in real time to increase the processing speed of the process.
[0032] In one embodiment, the resource requirements of the virtual node of the VNR include the number of instructions that need to be executed, and the resource requirements of the virtual link of the VNR include bandwidth.
[0033] Please refer to FIG. 3A to FIG. 3I , which is a schematic diagram of virtual network mapping in one embodiment of the present invention.
[0034] Figure 3AAn example of a physical network 300 including a plurality of physical nodes is shown, wherein the plurality of physical nodes are respectively labeled as A, B, C, and D, and the plurality of physical nodes A, B, C, and D are interconnected via physical links. The physical network 300 is managed by an infrastructure provider, and a virtual network can be established by the infrastructure provider on behalf of a service provider.
[0035] In this example, each physical node A, B, C and D is configured in an asymmetric mode, with a high-speed core processing capacity of 10 and a low-speed core processing capacity of 4. The service provider provides various services to users and can create virtual networks, each of which is used to provide customized services in a specific scenario. The virtual network includes multiple virtual network functions, called virtual nodes, which are connected via virtual links, and the virtual links require bandwidth resources. The service provider transmits each virtual network establishment requirement to the infrastructure provider with a VNR. In this embodiment, the mapping of the virtual network starts when the infrastructure provider's orchestrator receives a VNR310 with a low priority requirement from the service provider. In this example, the virtual network requested by the VNR310 has an AOV topology. The virtual network requested by the VNR310 includes three virtual nodes (labeled as a, b and c), and the three virtual nodes have a service order restriction, that is, virtual node a must be mapped first, followed by virtual node b and finally virtual node c. In addition, the numbers appearing in the boxes above the virtual nodes a, b, and c represent the resource requirements of the virtual nodes, respectively, and the numbers appearing next to the virtual links represent the resource requirements of the virtual links. In this embodiment, the infrastructure provider's orchestrator collects all input VNR requests in a time window, and then starts allocating resources in the next time window to meet the resource requirements. In this example, the time window is configured as a time slot, the VNR310 is received at t=T0 (t is the current time), and will be processed by the orchestrator in the next time slot (t=1T). The orchestrator maintains the current resource usage status of the entire physical network 300, calculates the solution for the VNR310 and reserves resources for it. FIG. 3B to FIG. 3H The reserved resources when the scheduler maps the VNR310 to the physical network 310 at each time slot are respectively shown.
[0036] Figure 3B The following is a schematic diagram of the resources reserved for the VNR 310 during t = 2T. As mentioned above, if there are available resources, the first virtual node of the low priority VNR can be mapped to a high-speed or standard-speed core when first mapped. In this example, virtual node a is mapped to the high-speed core of physical node D and occupies 6 units of resources.
[0037] Figure 3CThe diagram shows the resources reserved for the VNR310 during the period t=3T. In this embodiment, the transmission delay of the service via the virtual link is taken into account. Specifically, the processed instructions will encounter delays when transmitted via the physical link connecting two physical nodes. The transmission delay can be represented by a time slot. When a virtual link is mapped to a physical link, bandwidth and time slots must be reserved for the virtual link. The transmission delay can be calculated based on the number of megabytes generated after the instruction processing and the processing capacity of the physical link to which the virtual link is mapped. In this example, assuming that the transmission delay is 1 time slot, and the virtual link (a, b) is mapped to the physical link (D, A), the physical link (D, A) will reserve resources with a duration of 1 time slot and a bandwidth of 1 unit for the virtual link (a, b).
[0038] Figure 3D and Figure 3E The resources reserved for the VNR310 during t=4T and 5T are shown respectively. As mentioned above, the other virtual nodes of the low-priority VNR can only be mapped to the low-speed or standard-speed core for the first mapping. However, executing the instructions of the virtual node on the physical node will cause processing delays. Specifically, the processing delay can be calculated based on the resource requirements of the virtual node and the processing capacity of the physical node to which the virtual node is mapped. In this example, virtual node b can only be mapped to the low-speed core of physical node A. In this example, since the processing capacity of the low-speed core of physical node A is only 4 and the resource requirement of the virtual node is 5, when virtual node b is mapped to the low-speed core of physical node A, the processing delay is 2 time slots, and physical node A needs to reserve the resources of the low-speed core for virtual node b during the 2 time slots to complete the process of virtual node b.
[0039] Figure 3F The figure shows the resources reserved for the VNR 310 during t=6T and t=7T. In this example, the delayed transmission of the virtual link (b, c) is assumed to be 2 time slots, so when the virtual link is mapped to the physical link (A, B), the physical link (A, B) will reserve a resource with a duration of 2 time slots (6T and 7T respectively) and a bandwidth of 2 units for the virtual link (b, c).
[0040] Figure 3G and Figure 3H The resources reserved for the VNR 310 during t=8T and 9T are shown respectively. In this example, the virtual node c can only be mapped to the low-speed core of the physical node B. In this example, since the processing capacity of the low-speed core of the physical node B is only 4 and the resource requirement of the virtual node c is 5, when the virtual node c is mapped to the low-speed core of the physical node B, the processing delay is 2 time slots, and the physical node B needs to reserve the resources of the low-speed core for the virtual node c during the 2 time slots to complete the process of the virtual node c.
[0041] Fig. 3I Shown is the resource orchestration performed by the orchestrator for VNR310 when resources are available.
[0042] As mentioned above, the orchestrator receives the resource usage information reported by the computing node forwarded by the control node. After receiving the resource usage information, the orchestrator updates the resource usage status according to the resource usage information. Therefore, the orchestrator has the resource usage status of the physical network and can reallocate resources according to the resource usage status in each time slot. In this example, assuming that the physical network 300 currently only serves the VNR310, the orchestrator finds that the high-speed core of the physical node A has available resources during t=4T according to the resource usage status. Therefore, a new solution is calculated for the VNR310, and the control node that manages the physical node A is notified of the new solution, which migrates the instructions executed by the virtual node b mapped to the low-speed core of the physical node A to the high-speed core of the physical node A. According to the new solution, the processing delay of the virtual node b will be reduced to one time slot because the processing capacity of the high-speed core is sufficient, and the resources of the low-speed core originally reserved will be released at the same time. Therefore, the orchestrator can use the released resources to map other low-priority VNRs within the same period of time, which can increase the revenue of the service provider in the long run and also have the opportunity to increase the revenue of the infrastructure provider.
[0043] In addition to processing delay and transmission delay, propagation delay is also a key issue in the MEC network environment. In one embodiment, it is assumed that the geographical distance between two edge areas is far, for example, in different cities. Therefore, data transmitted through the physical link across the edge area will experience propagation delay. The propagation delay between two edge areas will be much higher than any delay in the same edge area. In one embodiment, the propagation delay can be calculated based on the geographical distance of the physical link connected to the edge area and the wave transmission speed of the physical link, and can be converted into time slot units. In another embodiment, the propagation delay can be obtained through a simple table lookup method. The main goal of the scheduler in solving the VNR solution, in addition to the effective use of physical network resources, is to minimize the total processing time of each VNR so that the infrastructure provider can accept more VNRs in the same time. In one embodiment, the total processing time of the VNR can be calculated based on the processing delay, transmission delay, and propagation delay.
[0044] Please refer to Figure 4 , which is a flowchart of a resource orchestration method in one embodiment of the present invention. The resource orchestration method can be executed by the orchestrator 102.
[0045] In step S402, the orchestrator 102 receives a virtual network request (VNR). The virtual network requested by the VNR has an activity-on-vertex (AOV) topology, which includes a plurality of virtual nodes connected via virtual links. After receiving the VNR, the orchestrator 102 stores the VNR in a queue.
[0046] Step S404, the scheduler 102 determines whether the time for periodic resource scheduling has expired. If the scheduler 102 determines that the time for periodic resource scheduling has expired, the coordinator reads VNRs from the queue in sequence (if more than one VNR is stored in the queue), and executes step S405 for each VNR; if the scheduler 102 determines that the time for periodic resource scheduling has not expired, the scheduler 102 continues to wait.
[0047] Step S406, the scheduler 102 calculates and solves a first virtual network mapping (VNE) solution for the VNR according to the resource requirements of the VNR, the current resource usage status, and the attributes of the physical nodes and physical links. The resource requirements of the VNR include priority, delay requirements, resource requirements of virtual nodes, and resource requirements of virtual links. The current resource usage status includes the available processing capacity of the physical nodes and the available processing capacity of the physical links.
[0048] In step S408, the orchestrator 102 determines whether there is a first VNE solution for the VNR. When the calculation result shows that there is no suitable VNE solution, the orchestrator 102 determines that the first VNE solution does not exist, and executes step S410 to reject the VNR. On the contrary, when the calculation result shows that there is a VNE solution, the orchestrator 102 determines that the first VNE solution exists, and executes step S412. In one embodiment, if the orchestrator 102 calculates and solves multiple VNR solutions, the VNR with the shortest total processing time is selected as the first VNE solution.
[0049] Step S412, the orchestrator 102 delivers the first VNE solution to at least one control node. Specifically, if the physical nodes mapped to the first VNE solution are located in different edge areas, the orchestrator 102 delivers the first VNE solution to different control nodes corresponding to different edge areas; if the physical nodes mapped to the first VNE solution are located in the same edge area, the orchestrator 102 delivers the first VNE solution to the control node corresponding to the edge area.
[0050] Please refer to Figure 5 , which is a flowchart of a resource scheduling method in another embodiment of the present invention, wherein the resource scheduling method may be performed by the scheduler 102 Figure 4 In this embodiment, each physical node in the physical network periodically reports information about resource usage to a corresponding control node, which then forwards the information to the scheduler 102.
[0051] Step S502: The scheduler 102 receives resource usage information, which is forwarded by the control node. The resource usage information includes the available processing capacity of the entity node and the available processing capacity of the entity link.
[0052] In step S504, the scheduler 102 updates the resource usage status according to the received resource usage information.
[0053] In step S506, the scheduler 102 calculates and solves a second VNE solution based on the resource demand of the VNR and the updated resource usage status, and determines whether there is a second VNE solution, wherein the total processing time of the VNR of the second VNE solution is shorter than the total processing time of the VNR of the first VNE solution. If it is determined that there is a second VNE solution, the scheduler 102 executes step S508; otherwise, the scheduler 102 executes step S510.
[0054] Step S508: The orchestrator 102 hands over the second VNE solution to at least one control node.
[0055] In step S510, the scheduler 102 determines whether the process of the VNR has been completed. If the scheduler 102 determines that the process of the VNR has not been completed, the scheduler 102 continues to execute step S502; otherwise, the scheduler 102 ends the process of the method.
[0056] Please refer to Figure 6, which is a schematic diagram of a resource orchestration device 600 in one embodiment of the present invention. The device 600 may be the orchestrator 102. The device 600 includes a processing unit 610 and a computer-readable storage medium 620. The processing unit 610 is electrically connected to the computer-readable storage medium 620. The processing unit 610 includes a microcontroller, a microprocessor or other circuits with computing processing capabilities, and executes or processes instructions, data and computer programs stored in the computer-readable storage medium 620. The computer-readable storage medium 620 includes a read-only memory (ROM), a random-access memory (RAM), a disk storage medium device, an optical storage medium device, a flash memory device, and an electrical, optical or other physical / tangible (e.g., non-temporary) storage device, etc. The computer-readable storage medium 620 is used to store one or more computer programs that control the operation of the device 600 and are executed by the processing unit 610. In this embodiment, the computer-readable storage medium 620 stores or encodes one or more computer programs and stores and configures parameters and data for execution by the processing unit 610. Figures 4 to 5 The method steps are shown.
[0057] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.
Claims
1. A resource orchestration method for a multi-access edge computing network, characterized in that: The multi-access edge computing network includes an orchestrator, a plurality of control nodes, a plurality of computing nodes, and a plurality of edge regions, wherein the orchestrator is used to provide real-time resource orchestration, including configuration of virtual and physical network functions, each of the plurality of control nodes is respectively configured to manage resources of computing nodes in a specific geographical location and physical link resources in an edge region, each of the plurality of edge regions is communicatively connected to each other via physical links, and the method is executed by the orchestrator, and the method includes: receiving a virtual network request, wherein the virtual network requested by the virtual network request has a vertex active topology, the resource requirements of the virtual network request include priority, latency requirements, resource requirements of virtual nodes, and resource requirements of virtual links, the resource requirements of the virtual nodes include the number of instructions to be executed, and the resource requirements of the virtual links include bandwidth; Determine whether the time for periodic resource scheduling has expired; When the time of periodic resource scheduling has arrived, a first virtual network mapping solution for solving the virtual network request is calculated according to the resource demand of the virtual network request, the resource usage status of the multi-access edge computing network, the attributes of the physical node, and the attributes of the physical link, wherein the attributes of the physical node include processing capacity and geographical area, and the processing capacity includes the number of instructions that can be completed per unit time, the attributes of the physical link include processing capacity and latency, and the processing capacity includes bandwidth; Determining whether the first virtual network mapping solution exists; and When it is determined that the first virtual network mapping solution exists, the first virtual network mapping solution is handed over to at least one control node of the multi-access edge computing network; otherwise, the virtual network request is rejected.
2. The resource arrangement method according to claim 1, characterized in that: The method further comprises: receiving resource usage information, wherein the resource usage information includes available processing capacity of the entity node and available processing capacity of the entity link; Updating the resource usage status of the multi-access edge computing network according to the received resource usage information; Calculate and solve a second virtual network mapping solution for the virtual network request according to the resource requirements of the virtual network request, the updated resource usage status of the multi-access edge computing network, the attributes of the physical nodes, and the attributes of the physical links; determining whether the second virtual network mapping solution exists; and When it is determined that the second virtual network mapping solution exists, the second virtual network mapping solution is handed over to at least one control node of the multi-access edge computing network.
3. The resource arrangement method according to claim 2, characterized in that: A total processing time of virtual network requests of the second virtual network mapping solution is shorter than a total processing time of virtual network requests of the first virtual network mapping solution.
4. A resource orchestration device for a multi-access edge computing network, characterized in that: The multi-access edge computing network further includes a plurality of control nodes, a plurality of computing nodes, and a plurality of edge regions. The resource orchestration device is used to provide real-time resource orchestration, including configuration of virtual and physical network functions. Each of the plurality of control nodes is respectively configured to manage resources of computing nodes in a specific geographical location and physical link resources in an edge region. Each of the plurality of edge regions is connected to each other in communication via physical links. The resource orchestration device includes: processing unit, and A computer-readable storage medium for storing at least one computer program, wherein the computer program comprises instructions executed by the processing unit, causing the processing unit to perform the following steps: receiving a virtual network request, wherein the virtual network requested by the virtual network request has a vertex active topology, the resource requirements of the virtual network request include priority, latency requirements, resource requirements of virtual nodes, and resource requirements of virtual links, the resource requirements of the virtual nodes include the number of instructions to be executed, and the resource requirements of the virtual links include bandwidth; Determine whether the time for periodic resource scheduling has expired; When the time of periodic resource scheduling has arrived, a first virtual network mapping solution for solving the virtual network request is calculated according to the resource demand of the virtual network request, the resource usage status of the multi-access edge computing network, the attributes of the physical node, and the attributes of the physical link, wherein the attributes of the physical node include processing capacity and geographical area, and the processing capacity includes the number of instructions that can be completed per unit time, the attributes of the physical link include processing capacity and latency, and the processing capacity includes bandwidth; Determining whether the first virtual network mapping solution exists; and When it is determined that the first virtual network mapping solution exists, the first virtual network mapping solution is handed over to at least one control node of the multi-access edge computing network; otherwise, the virtual network request is rejected.
5. The resource arrangement device according to claim 4, characterized in that: The steps include: receiving resource usage information, wherein the resource usage information includes available processing capacity of the entity node and available processing capacity of the entity link; Updating the resource usage status of the multi-access edge computing network according to the received resource usage information; Calculate and solve a second virtual network mapping solution for the virtual network request according to the resource requirements of the virtual network request, the updated resource usage status of the multi-access edge computing network, the attributes of the physical nodes, and the attributes of the physical links; determining whether the second virtual network mapping solution exists; and When it is determined that the second virtual network mapping solution exists, the second virtual network mapping solution is handed over to at least one control node of the multi-access edge computing network.
Citation Information
Patent Citations
Collaborative virtual network mapping method
CN105939244A
Virtual network mapping method of elastic optical network facing SDN
CN106059799A
Cross-domain deployment method for network service function chain
CN107395501A
Distributed virtual network embedding
US20180227243A1
Efficient message forwarding for distributed resource orchestration
US20180227244A1
