A cloud edge data center network service offloading protection method and readable storage medium

Abstract

The present invention relates to a cloud edge data center network service offloading protection method, comprising initializing the cloud edge data center, obtaining the network service to be offloaded of the cloud edge data center; preferentially offloading the network service to be offloaded in the edge computing area, and if there is no node that meets the demand in the edge computing area, transferring it to the cloud computing area for offloading; in the edge computing area, obtaining an edge node that meets the computing resource demand of the service to be offloaded as a target edge offloading node, constructing an optimal working path that meets the bandwidth resource demand and an optimal protection path that does not intersect with the corresponding working link to form a redundant working path combination; if the optimal working path in the redundant working path combination has no fault or congestion, the optimal working path is used as the current working path, otherwise the optimal protection path is used as the current working path, and the network service to be offloaded is sent to the target edge offloading node via the current working path for offloading.

Description

Technical Field

[0001] The present invention relates to the field of cloud edge data center network technology, and in particular to a cloud edge data center network service offloading protection method and a computer-readable storage medium. Background Art

[0002] In the era of cloud-edge data center networks, with the expansion of the number and scale of data centers, the development needs of application businesses such as large models, big data, and AI technologies are driving changes in the form of networks and computing networks, and the network structure has become extremely complex.

[0003] Due to the scale and complexity of cloud-edge data center networks, network control and management have become increasingly difficult. Different business scenarios have different requirements for network performance, latency, bandwidth, etc., which further increases the difficulty of network management. The complex network structure increases the probability of failure. Failure in any link may trigger a chain reaction, leading to business interruption.

[0004] When a network failure occurs in the cloud-edge data center network and network services are interrupted, the transmission of network services will be unreliable, which will cause various losses and may also cause business security issues such as data inconsistency. The network services to be offloaded cannot be transmitted to the target node in a timely and accurate manner for unloading, reducing the business offloading efficiency and network resource utilization. Summary of the invention

[0005] To this end, the technical problem to be solved by the present invention is to overcome the problem in the prior art that due to the complexity of the cloud-edge data center network structure, network failures are prone to occur, resulting in interruption of network service transmission and inability to efficiently offload network services.

[0006] In order to solve the above technical problems, the present invention provides a cloud edge data center network service offloading protection method, comprising:

[0007] Initialize the cloud edge data center and collect the node computing resource occupancy status and bandwidth resource occupancy status of all edge nodes in the edge computing area and all cloud nodes in the cloud computing area in real time;

[0008] Obtain the network services to be offloaded from the cloud edge data center, as well as the source nodes, computing resource requirements, and bandwidth resource requirements of the network services to be offloaded;

[0009] Using the master-slave network control method of the edge computing area and the cloud computing area, based on the node computing resource occupancy status of each edge node in the edge computing area where the task to be offloaded is located, it is determined whether there is an edge node that meets the computing resource requirements of the network service to be offloaded:

[0010] If it exists, the edge node is used as the target edge offloading node, and the network active and standby controller is used to obtain K candidate shortest working paths and K candidate shortest protection paths that do not intersect with the working links of the K candidate shortest working paths based on the K shortest path algorithm;

[0011] Select the candidate shortest working path with the least number of hops from the K candidate shortest working paths as the optimal working path, and determine whether the bandwidth of the optimal working path meets the bandwidth resource demand of the network service to be offloaded based on the bandwidth resource occupancy status:

[0012] If the condition is satisfied, the candidate shortest protection path with the least number of hops is selected from the K candidate shortest protection paths as the optimal protection path, and the optimal protection path is determined based on the bandwidth resource occupancy status.

[0013] Whether the bandwidth of the path meets the bandwidth resource requirements of the network service to be offloaded:

[0014] If the conditions are met, the optimal protection path and its corresponding optimal working path are combined into a redundant working path combination of the network service to be offloaded;

[0015] If not, then traverse the K candidate shortest protection paths in order from small to large according to the number of hops in the path until the candidate shortest protection path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded is obtained as the optimal protection path, and the optimal protection path and its corresponding optimal working path are combined into a redundant working path combination of the network service to be offloaded;

[0016] If not, then traverse the K candidate shortest working paths in sequence until a candidate shortest working path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded is obtained as the optimal working path, and the optimal protection path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded is selected from the K candidate shortest protection paths for the optimal working path to form a redundant working path combination for the network service to be offloaded;

[0017] The optimal working path in the redundant working path combination of the network service to be offloaded is used as the current working path, and the network service to be offloaded is sent to the target edge offloading node for offloading; if the optimal working path in the redundant working path combination fails or is congested, the optimal protection path corresponding to the optimal working path is used as the current working path, and the network service to be offloaded is sent to the target edge offloading node through the current working path for offloading;

[0018] If it does not exist, based on the node computing resource occupancy status and bandwidth resource occupancy status of all cloud nodes in the cloud computing area, a cloud node that meets the computing resource requirements and bandwidth resource requirements of the network service to be offloaded is selected as the target cloud offloading node; using the K shortest path algorithm, K candidate shortest working paths from the source node to the target cloud offloading node are obtained, and the candidate shortest working path with the least hops is selected as the current working path, and the network service to be offloaded is sent to the target cloud offloading node for offloading.

[0019] Preferably, if it is judged based on the bandwidth resource occupancy status that the bandwidth of the optimal protection path does not meet the bandwidth resource requirements of the network service to be unloaded, and K candidate shortest protection paths are traversed in sequence, and no candidate shortest protection path whose bandwidth meets the bandwidth resource requirements of the network service to be unloaded can be obtained, then the unloading of the network service to be unloaded fails.

[0020] Preferably, if it is judged based on the bandwidth resource occupancy status that the bandwidth of the optimal working path does not meet the bandwidth resource requirements of the network service to be unloaded, and K candidate shortest working paths are traversed in sequence, and no candidate shortest working path whose bandwidth meets the bandwidth resource requirements of the network service to be unloaded can be obtained, then the unloading of the network service to be unloaded fails.

[0021] Preferably, based on the node computing resource occupancy status and bandwidth resource occupancy status of all cloud nodes in the cloud computing area, a cloud node that meets the computing resource requirements and bandwidth resource requirements of the network service to be offloaded is selected. If no cloud node exists, the offloading of the network service to be offloaded fails.

[0022] Preferably, the initializing the cloud edge data center includes:

[0023] Initialize the edge computing area and configure edge links for each edge node in the edge computing area;

[0024] Initialize the cloud computing area and configure cloud links for each cloud node in the cloud computing area.

[0025] Preferably, both the edge link and the cloud link are optical fiber links.

[0026] Preferably, obtaining the network services to be unloaded in the cloud edge data center includes: obtaining the computing resource requirements, bandwidth resource requirements, source nodes, unloading nodes, arrival time, duration and departure time of the network services to be unloaded.

[0027] Preferably, after the network service to be uninstalled is uninstalled, the method further includes: releasing the computing resources and bandwidth resources occupied by the network service to be uninstalled, and removing the route of the current working path occupied by the network service to be uninstalled.

[0028] Preferably, unloading the network service to be unloaded includes: transmitting the network task to be unloaded to a target unloading node via the current working path, and executing the network service to be unloaded on the target unloading node; the target unloading node includes a target edge unloading node and a target cloud unloading node.

[0029] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the cloud edge data center network service offloading protection method as described above are implemented.

[0030] The above technical solution of the present invention has the following beneficial effects compared with the prior art:

[0031] The cloud edge data center network service offloading protection method described in the present invention selects K candidate shortest working paths that meet the requirements and K candidate shortest protection paths that do not intersect with the links of the K candidate shortest working paths for the network service to be offloaded in the edge computing area of ​​the cloud edge data center based on the node computing resource occupancy status and bandwidth resource occupancy status of the edge nodes; selects the optimal working path and the optimal protection path to form a redundant working path combination; flexibly calls the optimal working path or the optimal candidate path to flexibly allocate network resources for the network service to be offloaded, and when the optimal working path fails, uses the optimal protection path to improve the fault recovery capability of the network service offloading, solves the problem of network service loss caused by network failure, ensures efficient offloading of network services, improves the transmission reliability of network services, and thus improves the survivability of the cloud edge data center network.

[0032] In addition, according to the resource characteristics of multiple areas of the cloud edge data center, the present invention preferentially unloads the network services to be unloaded to the edge computing area; if no path that meets the needs of the network services to be unloaded can be found in the edge computing area, a network service transfer method between the edge and cloud computing areas is introduced, and a unified and joint resource transfer and unloading method between the edge computing area and the cloud computing area is adopted to transfer the network services to be unloaded to the cloud computing area for unloading; by intelligently identifying network services, work resources are flexibly allocated to network services, the service access unloading protection capability of the cloud edge data center is improved, and the resource utilization of the cloud edge data center is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to make the content of the present invention more clearly understood, the present invention is further described in detail below according to specific embodiments of the present invention in conjunction with the accompanying drawings, wherein:

[0034] Figure 1 It is a flowchart of the steps of the cloud edge data center network service offloading protection method provided by the present invention;

[0035] Figure 2It is a flowchart of the specific steps of the cloud edge data center network service offloading protection method provided by the present invention;

[0036] Figure 3 It is a schematic diagram of the cloud edge data center network structure provided by the present invention;

[0037] Figure 4 This is a structural diagram of the cloud edge data center network service offloading system provided by the present invention. DETAILED DESCRIPTION

[0038] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0039] Reference Figure 1 As shown, the step flow chart of the cloud edge data center network service offloading protection method of the present invention includes:

[0040] Initialize the cloud edge data center and collect the node computing resource occupancy status and bandwidth resource occupancy status of all edge nodes in the edge computing area and all cloud nodes in the cloud computing area in real time;

[0041] Obtain the network services to be offloaded from the cloud edge data center, as well as the source nodes, computing resource requirements, and bandwidth resource requirements of the network services to be offloaded;

[0042] Using the master-slave network control method of the edge computing area and the cloud computing area, based on the node computing resource occupancy status of each edge node in the edge computing area where the task to be offloaded is located, it is determined whether there is an edge node that meets the computing resource requirements of the network service to be offloaded:

[0043] If it exists, the edge node is used as the target edge offloading node, and the network active and standby controller is used to obtain K candidate shortest working paths and K candidate shortest protection paths that do not intersect with the working links of the K candidate shortest working paths based on the K shortest path algorithm;

[0044] Select the candidate shortest working path with the least number of hops from the K candidate shortest working paths as the optimal working path, and determine whether the bandwidth of the optimal working path meets the bandwidth resource demand of the network service to be offloaded based on the bandwidth resource occupancy status:

[0045] If the condition is satisfied, the candidate shortest protection path with the least number of hops is selected from the K candidate shortest protection paths as the optimal protection path, and the optimal protection path is determined based on the bandwidth resource occupancy status.

[0046] Whether the bandwidth of the path meets the bandwidth resource requirements of the network service to be offloaded:

[0047] If the conditions are met, the optimal protection path and its corresponding optimal working path are combined into a redundant working path combination of the network service to be offloaded;

[0048] If not, then traverse the K candidate shortest protection paths in order from small to large according to the number of hops in the path until the candidate shortest protection path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded is obtained as the optimal protection path, and the optimal protection path and its corresponding optimal working path are combined into a redundant working path combination of the network service to be offloaded;

[0049] If not, then traverse the K candidate shortest working paths in sequence until a candidate shortest working path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded is obtained as the optimal working path, and the optimal protection path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded is selected from the K candidate shortest protection paths for the optimal working path to form a redundant working path combination for the network service to be offloaded;

[0050] The optimal working path in the redundant working path combination of the network service to be offloaded is used as the current working path, and the network service to be offloaded is sent to the target edge offloading node for offloading; if the optimal working path in the redundant working path combination fails or is congested, the optimal protection path corresponding to the optimal working path is used as the current working path, and the network service to be offloaded is sent to the target edge offloading node through the current working path for offloading;

[0051] If it does not exist, based on the node computing resource occupancy status and bandwidth resource occupancy status of all cloud nodes in the cloud computing area, a cloud node that meets the computing resource requirements and bandwidth resource requirements of the network service to be offloaded is selected as the target cloud offloading node; using the K shortest path algorithm, K candidate shortest working paths from the source node to the target cloud offloading node are obtained, and the candidate shortest working path with the least hops is selected as the current working path, and the network service to be offloaded is sent to the target cloud offloading node for offloading.

[0052] Based on the above embodiment, when obtaining a node that meets the computing resource requirement or the bandwidth resource requirement at each step, if the node cannot be obtained, it is considered that the unloading of the network service to be unloaded has failed. The specific situations include:

[0053] ① If it is determined based on the bandwidth resource occupancy status that the bandwidth of the optimal protection path does not meet the bandwidth resource requirements of the network service to be offloaded, and the K candidate shortest protection paths are traversed in sequence, and no candidate shortest protection path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded can be obtained, then the offloading of the network service to be offloaded fails.

[0054] ② If it is determined based on the bandwidth resource occupancy status that the bandwidth of the optimal working path does not meet the bandwidth resource requirements of the network service to be offloaded, and the K candidate shortest working paths are traversed in sequence, and no candidate shortest working path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded can be obtained, then the offloading of the network service to be offloaded fails.

[0055] ③ Based on the node computing resource occupancy status and bandwidth resource occupancy status of all cloud nodes in the cloud computing area, a cloud node that meets the computing resource requirements and bandwidth resource requirements of the network service to be offloaded is selected. If no cloud node exists, the offloading of the network service to be offloaded fails.

[0056] Specifically, the initialization of the cloud edge data center includes: initializing the edge computing area and configuring edge links for each edge node in the edge computing area; initializing the cloud computing area and configuring cloud links for each cloud node in the cloud computing area. The edge links and cloud links are both optical fiber links.

[0057] Specifically, obtaining the network services to be unloaded in the cloud edge data center includes: obtaining the computing resource requirements, bandwidth resource requirements, source nodes, unloading nodes, arrival time, duration and departure time of the network services to be unloaded. Unloading the network services to be unloaded includes: transmitting the network tasks to be unloaded to the target unloading node via the current working path, and executing the network services to be unloaded on the target unloading node; the target unloading nodes include the target edge unloading nodes and the target cloud unloading nodes. After unloading the network services to be unloaded, it also includes: releasing the computing resources and bandwidth resources occupied by the network services to be unloaded, and dismantling the routes of the current working path occupied by the network services to be unloaded.

[0058] The present invention considers cloud computing and edge computing as the computing power provider of network services, uses optical networks as cloud edge data center connection carriers, provides large bandwidth access capabilities, and solves the problem of high reliability protection of cloud edge data center network services through the working and protection path mechanism of network services, meeting the application business requirements of strong computing power, wide connection, low latency, and high survivability of the optical network of the cloud edge data center. If the edge computing area can meet the computing resource requirements of the service, the network service is preferentially unloaded to the edge computing area, and the network service is protected in the edge area to ensure the survivability of the network service. If the network resource requirements that meet the network service cannot be found in the edge computing area, a unified and joint resource transfer and unloading method of the edge computing area and the cloud computing area is adopted, and the network service needs to be transferred to the cloud computing area for unloading, and the network service is protected in the cloud area at the same time. In this way, according to the bandwidth resources of the network service, the bandwidth resources of the network are found and allocated in the protection path where the working path of the network and the link do not intersect, so as to achieve the goal of successful unloading and protection of the network service. At the same time, the present invention takes into account the failure problem of the cloud edge data center network, adopts a network service unloading protection method, protects each network service, solves the loss problem of network services caused by network failures, and improves the fault recovery capability of network service unloading. By introducing a professional protection path for services, adopting the control method of the primary and standby network controllers, and utilizing the unified and joint resource unloading protection method of the cloud edge data center, the control and protection capability of network services is improved, the unloading protection method of the cloud edge data center network services is realized, and the problem of efficient unloading of the cloud edge data center network services is solved.

[0059] Specifically, considering the failure problem of the cloud edge data center network, the embodiment of the present invention adopts a network service unloading protection method to protect each network service, solve the problem of network service loss caused by network failure, and improve the fault recovery capability of network service unloading. By introducing a professional protection path method for the service, adopting the control method of the primary and standby network controllers, and utilizing the unified and joint resource unloading protection method of the cloud edge data center, the control and protection capabilities of the network service are improved, and the unloading protection method of the cloud edge data center network service is implemented to solve the problem of efficient unloading of the cloud edge data center network service. The present invention introduces an unloading protection method for the edge and cloud computing areas based on the resource characteristics of the cloud edge data center network, and utilizes the cloud edge data center network controller to intelligently identify the network service and accurately calculate the protection path with non-intersecting work and links, providing a flexible allocation of working resources and protection resource protection method for network services, solving the cloud edge data center service access unloading protection capability, and improving the resource efficiency of the cloud edge data center network. Specifically, refer to Figure 2 As shown, it is a flowchart of the specific steps of the cloud edge data center network service offloading protection method, and the specific steps include:

[0060] Step 1: Generate a set of cloud edge data center network services and initialize the cloud edge data center network, including the node computing resources of the cloud edge data center and the links connected by optical fiber.

[0061] Step 2: Based on the offloading requirements of the cloud edge data center network services, combined with the resource attributes of the cloud edge data center, and based on the computing resource characteristics of the edge data center and the cloud computing data center, the primary and secondary network control methods of the edge computing area and the cloud computing area are adopted to determine whether the cloud edge data center network services have sufficient computing resources in all edge computing areas. If there are sufficient computing resources in all edge computing areas, execute step 3; if not, execute step 7;

[0062] Step 3: In order to offload each network service, the joint control mode of the primary and backup networks in the edge computing area and the cloud area is used, and the K shortest path algorithm is adopted to establish K candidate shortest working paths in the edge computing area, and the path with the smallest number of hops is preferentially selected as the preferred working path; if the preferred working path has no available bandwidth resources, it is necessary to traverse all K candidate working paths until the last working path is found, so as to minimize the network bandwidth resource usage;

[0063] Step 4: Using the joint control mode of the primary and backup networks in the edge computing area and the cloud area, adopt the K shortest path algorithm to establish K candidate shortest protection paths in the edge computing area, and give priority to the path with the smallest number of hops as the preferred protection path; if the preferred path has no available bandwidth resources, it is necessary to traverse all K candidate link-disjoint protection paths until the last link-disjoint protection path is found, which can minimize the network bandwidth resource usage;

[0064] Step 5: Find the bandwidth resource requirements that meet the network service in the selected working path and the link-disjoint protection path. If both the working path and the link-disjoint protection path meet the bandwidth requirements of the service, the network service is successfully unloaded and the network operation ends.

[0065] Step 6: If computing resources and bandwidth resources that meet the needs of network services cannot be found in the edge computing area, the network resource transfer method between edge computing and cloud computing areas is adopted. It is necessary to transfer network services from the edge computing area to the cloud computing area for unloading, and perform unloading protection of network services in the cloud computing area, and execute step 7.

[0066] Step 7: If the cloud computing area has sufficient computing resources and bandwidth resources, computing resources, working path bandwidth resources, and protection path bandwidth resources are allocated to the network service, and the network service can be successfully offloaded to the cloud computing area, thus achieving successful offloading of the network service; if there are not sufficient computing resources and bandwidth resources, the cloud edge data center network service cannot be successfully offloaded, and the network service offloading fails.

[0067] Based on the above embodiments, in the embodiments of the present invention, in order to establish a cloud edge data center network service, according to the characteristics of the cloud edge data center network resources, first determine whether the cloud edge data center can provide sufficient computing resources, and use the K conditional shortest path method to calculate K candidate working paths and link-disjoint protection paths for the network service, and search and allocate network bandwidth resources on the selected working paths and link-disjoint protection paths, so as to establish the network service. Figure 3 As shown, it is a schematic diagram of the cloud edge data center network structure of an embodiment of the present invention; specifically, the embodiment of the present invention is based on a cloud edge data center network consisting of two edge computing areas I and II and a cloud computing area, and performs network center service offloading protection, and the specific steps include:

[0068] S201: Initialize the cloud edge data center network and generate two network services CR1 (I1, d, 6, 15, 400 Gbps) and CR2 (II5, d, 5, 7, 120 Gbps).

[0069] CR1(I1,d,6,15,400Gbps) indicates that the network service is unloaded from the edge node I1 to a node d in the network. The computing resource requirements of the edge node and the unloading node are 6 computing units and 15 computing units respectively, and the bandwidth resource required by the service is 400Gbps.

[0070] CR2(II5,d,5,7,120Gbps) indicates offloading from edge node II5 to a node d in the network. The computing resource requirements of the edge node and the offloading node are 5 computing units and 7 computing units respectively, and the bandwidth resource required by the business is 120Gbps.

[0071] S202: For network service CR1 (I1, d, 6, 15, 400 Gbps), using the cloud edge data center network active and standby controller, in edge computing area I, the computing resources provided from edge computing node I1 to any edge offloading node, such as edge computing nodes I2, I3, I4, and I5, cannot meet the computing resource requirements of network service CR1 (I1, d, 6, 15, 400 Gbps). In this way, working paths p1, p2, p3, and p4 and link-disjoint protection paths b1, b2, b3, and b4 cannot be successfully established.

[0072] S203: Using the network controller, the network service CR1 (I1, d, 6, 15, 400 Gbps) needs to be transferred to the cloud computing area for unloading. In this way, the network service unloading node d can be unloaded to the node D in the cloud computing area.

[0073] S204: Use the K shortest path algorithm (K=1) to calculate the working path and the link-disjoint protection path from the edge node I1 to the cloud computing area node D. The working path is p5 and the link-disjoint protection path is b5. That is, the nodes passed by the working path are I1, I5, I4, D and the nodes of the link-disjoint protection path are I1, I2, A, D.

[0074] S205: On the working path p5 and the link-disjoint protection path b5, find and allocate the bandwidth requirement of the service CR1 (I1, d, 6, 15, 400 Gbps). If the bandwidth meets the requirement, the network service can be successfully established.

[0075] Similarly, for network service CR2 (II5, d, 5, 7, 120 Gbps), the cloud edge data center network master and standby controller is used to unload from edge computing node II5 to any edge unloading node in edge computing area II. If II2 is selected as the network service unloading node d; the K shortest path algorithm (K = 1) is used to calculate the working path and the link-disjoint protection path from edge node II5 to edge node II2. The working path is p6 and the link-disjoint protection path is b6, that is, the working path is p6, the nodes passed are II5, II1, and II2, the link-disjoint protection path is b6, and the node passed is I I5, II3, II2; on this working path p6 and the link-disjoint protection path b6, find and allocate the bandwidth requirement of service CR2 (II5, d, 5, 7, 120 Gbps). If the bandwidth meets the requirement, network service CR2 (II5, d, 5, 7, 120 Gbps) can be successfully established; if the network services CR1 (I1, d, 6, 15, 400 Gbps) and CR2 (II5, d, 5, 7, 120 Gbps) are completed, release the computing resources and bandwidth resources occupied by the network, and dismantle the established working paths p5 and p6, as well as the link-disjoint protection paths b5 and b6.

[0076] Therefore, the present invention mainly aims at the problem of service offloading protection of cloud edge data center network, proposes a method and system for service offloading protection of cloud edge data center network, and solves the problem of flexible allocation and scheduling of cloud edge data center network services. According to the resource characteristics of multiple regions of cloud edge data center, the present invention introduces the transfer method of edge and cloud computing regions, solves the problem of service access protection capability of cloud edge data center, and improves the resource efficiency of cloud edge data center network.

[0077] The present invention mainly aims at the problem of service offloading in cloud edge data center network, proposes a cloud edge data center network service offloading protection method and system, and solves the problem of flexible allocation and scheduling of cloud edge data center network services. The present invention preferentially offloads network services to the edge computing area according to the resource characteristics of multiple areas of the cloud edge data center. If the network working and protection resource requirements that meet the network service cannot be found in the edge computing area, a network service transfer method between the edge and cloud computing areas is introduced, and a unified and joint resource transfer and offloading method between the edge computing area and the cloud computing area is adopted. The network service needs to be transferred to the cloud computing area for offloading, and a protection path with non-intersecting working paths and links is established, as well as working resources and protection resources are allocated. A method is provided for flexibly allocating network services to network services, which solves the service access capability of the cloud edge data center, achieves the goal of successfully offloading network services, and improves the survivability and resource efficiency of the cloud edge data center network. Specifically, based on the above embodiments, with reference to Figure 4 As shown, the embodiment of the present invention also provides a cloud edge data center network service offloading system structure diagram, which specifically includes:

[0078] (1) Network service generation module, which mainly includes information such as computing resources, bandwidth resources, source nodes, unloading nodes, arrival time, duration, and departure time of the service.

[0079] (2) Network initialization module: This module is mainly used to initialize the edge network and cloud computing network, including the initialization of computing resources and bandwidth resources such as edge nodes, edge links, cloud computing nodes, and cloud computing links.

[0080] (3) Working path and link-disjoint protection path calculation module: This module mainly uses K shortest path calculations to provide transmission working paths and link-disjoint protection paths for network service offloading. Among the selected candidate working paths and link-disjoint protection paths, it searches and allocates the required bandwidth resources for network services.

[0081] (4) Network judgment module: This module mainly judges whether the edge computing area and cloud computing area of ​​the cloud-edge data center network have sufficient computing resources and bandwidth resources required for network services, providing a judgment basis for establishing network services.

[0082] (5) Working resource and protection resource allocation module, which mainly allocates computing resources, working bandwidth resources, and protection bandwidth resources for network services.

[0083] (6) Network master-slave control module: This module can globally and in real time perceive the node computing resource occupancy status of the cloud-edge data center network, evaluate the bandwidth resource occupancy status of the network link, and monitor the faults of the network's edge nodes, cloud computing nodes, and optical fiber links.

[0084] (7) Network resource collaboration module: This module mainly controls, manages, transfers, and schedules the services, computing resources, and bandwidth resources of the cloud-edge data center network.

[0085] (8) Network service transfer module: This module mainly provides transfer and unloading functions for network services. If no network service can be unloaded in all edge computing areas, the network service can be transferred to the cloud computing area of ​​the network for unloading.

[0086] (9) Network resource status monitoring module: This module mainly monitors the computing resources, working bandwidth resources, and protection bandwidth resources of the cloud-edge data center network in real time. It can monitor the network resource usage information in real time, which is conducive to the rapid unloading of network services and the protection and recovery of network services.

[0087] (10) Release and removal module: After completing the data transmission of network services, this module releases the computing resources and bandwidth resources of the cloud-edge data center network and removes the network routing.

[0088] The embodiment of the present invention completes the unloading of cloud-edge data center network services through mutual collaboration between different modules. In addition to the above-mentioned modules, the embodiment of the present invention also includes a coordination and early warning module between different modules, which is used to perform coordination functions between various modules to ensure the successful establishment and execution of cloud-edge data center network services.

[0089] The cloud edge data center network service offloading protection method described in the present invention selects K candidate shortest working paths that meet the requirements and K candidate shortest protection paths that do not intersect with the links of the K candidate shortest working paths for the network service to be offloaded in the edge computing area of ​​the cloud edge data center based on the node computing resource occupancy status and bandwidth resource occupancy status of the edge nodes; selects the optimal working path and the optimal protection path to form a redundant working path combination; flexibly calls the optimal working path or the optimal candidate path to flexibly allocate network resources for the network service to be offloaded, and when the optimal working path fails, uses the optimal protection path to improve the fault recovery capability of the network service offloading, solves the problem of network service loss caused by network failure, ensures efficient offloading of network services, improves the transmission reliability of network services, and thus improves the survivability of the cloud edge data center network. In addition, according to the resource characteristics of multiple areas of the cloud edge data center, the present invention preferentially unloads the network services to be unloaded to the edge computing area; if no path that meets the needs of the network services to be unloaded can be found in the edge computing area, a network service transfer method between the edge and cloud computing areas is introduced, and a unified and joint resource transfer and unloading method between the edge computing area and the cloud computing area is adopted to transfer the network services to be unloaded to the cloud computing area for unloading; by intelligently identifying network services, work resources are flexibly allocated to network services, the service access unloading protection capability of the cloud edge data center is improved, and the resource utilization of the cloud edge data center is improved.

[0090] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0091] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes 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 generate 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.

[0092] 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.

[0093] 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.

[0094] Obviously, the above embodiments are merely examples for the purpose of clear explanation and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the present invention.

Claims

1. A cloud edge data center network service offloading protection method, characterized in that: include: Initialize the cloud edge data center and collect the node computing resource occupancy status and bandwidth resource occupancy status of all edge nodes in the edge computing area and all cloud nodes in the cloud computing area in real time; Obtain the network services to be offloaded from the cloud edge data center, as well as the source nodes, computing resource requirements, and bandwidth resource requirements of the network services to be offloaded; Using the master-slave network control method of the edge computing area and the cloud computing area, based on the node computing resource occupancy status of each edge node in the edge computing area where the task to be offloaded is located, it is determined whether there is an edge node that meets the computing resource requirements of the network service to be offloaded: If it exists, the edge node is used as the target edge offloading node, and the network active and standby controller is used to obtain K candidate shortest working paths and K candidate shortest protection paths that do not intersect with the working links of the K candidate shortest working paths based on the K shortest path algorithm; Select the candidate shortest working path with the least number of hops from the K candidate shortest working paths as the optimal working path, and determine whether the bandwidth of the optimal working path meets the bandwidth resource demand of the network service to be offloaded based on the bandwidth resource occupancy status: If so, the candidate shortest protection path with the least number of hops is selected from the K candidate shortest protection paths as the optimal protection path, and based on the bandwidth resource occupancy status, it is determined whether the bandwidth of the optimal protection path meets the bandwidth resource requirements of the network service to be offloaded: If the conditions are met, the optimal protection path and its corresponding optimal working path are combined into a redundant working path combination of the network service to be offloaded; If not, then traverse the K candidate shortest protection paths in order from small to large according to the number of hops in the path until the candidate shortest protection path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded is obtained as the optimal protection path, and the optimal protection path and its corresponding optimal working path are combined into a redundant working path combination of the network service to be offloaded; If not, then traverse the K candidate shortest working paths in sequence until a candidate shortest working path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded is obtained as the optimal working path, and the optimal protection path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded is selected from the K candidate shortest protection paths for the optimal working path to form a redundant working path combination for the network service to be offloaded; The optimal working path in the redundant working path combination of the network service to be offloaded is used as the current working path, and the network service to be offloaded is sent to the target edge offloading node for offloading; If the optimal working path in the redundant working path combination fails or is congested, the optimal protection path corresponding to the optimal working path is used as the current working path, and the network service to be offloaded is sent to the target edge offloading node through the current working path for offloading; If it does not exist, based on the node computing resource occupancy status and bandwidth resource occupancy status of all cloud nodes in the cloud computing area, a cloud node that meets the computing resource requirements and bandwidth resource requirements of the network service to be offloaded is selected as the target cloud offloading node; using the K shortest path algorithm, K candidate shortest working paths from the source node to the target cloud offloading node are obtained, and the candidate shortest working path with the least hops is selected as the current working path, and the network service to be offloaded is sent to the target cloud offloading node for offloading.

2. The cloud edge data center network service offloading protection method according to claim 1 is characterized in that: If it is determined based on the bandwidth resource occupancy status that the bandwidth of the optimal protection path does not meet the bandwidth resource requirements of the network service to be offloaded, and the K candidate shortest protection paths are traversed in sequence but no candidate shortest protection path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded can be obtained, then the offloading of the network service to be offloaded fails.

3. The cloud edge data center network service offloading protection method according to claim 1 is characterized in that: If it is determined based on the bandwidth resource occupancy status that the bandwidth of the optimal working path does not meet the bandwidth resource requirements of the network service to be offloaded, and the K candidate shortest working paths are traversed in sequence but no candidate shortest working path whose bandwidth meets the bandwidth resource requirements of the network service to be offloaded can be obtained, then the offloading of the network service to be offloaded fails.

4. The cloud edge data center network service offloading protection method according to claim 1 is characterized in that: Based on the node computing resource occupancy status and bandwidth resource occupancy status of all cloud nodes in the cloud computing area, a cloud node that meets the computing resource requirements and bandwidth resource requirements of the network service to be offloaded is selected. If no cloud node exists, the offloading of the network service to be offloaded fails.

5. The cloud edge data center network service offloading protection method according to claim 1 is characterized in that: The initialization of the cloud edge data center includes: Initialize the edge computing area and configure edge links for each edge node in the edge computing area; Initialize the cloud computing area and configure cloud links for each cloud node in the cloud computing area.

6. The cloud edge data center network service offloading protection method according to claim 5 is characterized in that: The edge link and the cloud link both use optical fiber links.

7. The cloud edge data center network service offloading protection method according to claim 1 is characterized in that: Obtain the network services to be unloaded in the cloud edge data center, including: obtaining the computing resource requirements, bandwidth resource requirements, source nodes, unloading nodes, arrival time, duration and departure time of the network services to be unloaded.

8. The cloud edge data center network service offloading protection method according to claim 1 is characterized in that: After the network service to be uninstalled is uninstalled, the method further includes: releasing the computing resources and bandwidth resources occupied by the network service to be uninstalled, and removing the route of the current working path occupied by the network service to be uninstalled.

9. The cloud edge data center network service offloading protection method according to claim 1 is characterized in that: Unloading the network service to be unloaded includes: transmitting the network task to be unloaded to the target unloading node via the current working path, and executing the network service to be unloaded on the target unloading node; the target unloading node includes a target edge unloading node and a target cloud unloading node.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the cloud edge data center network service offloading protection method as described in any one of claims 1 to 9 are implemented.

Images