Computing power renting method and device, equipment, medium and data processing system

Abstract

The present disclosure provides a computing power leasing method, device, equipment, medium and data processing system, relating to the technical field of data processing. The method comprises: a central scheduling system receiving a computing power leasing request from an edge node, the computing power leasing request comprising an address of the edge node, leased computing power resource information, leasing time information and a leasing cost threshold; calculating a leasing cost corresponding to each node according to a preset network topology map, the address of the edge node and the leased computing power resource information; determining a plurality of alternative nodes based on the leasing cost and the leasing cost threshold; calculating a similarity value of the alternative nodes using data within a preset time and the edge node demand data; and determining a target leasing node from the alternative nodes according to the leasing cost and the similarity value of each alternative node. According to the embodiment of the present disclosure, the problem of tight computing power of the edge node can be alleviated.

Description

Technical Field

[0001] This disclosure relates to the field of data processing technology, and in particular to a method, apparatus, equipment, medium and data processing system for renting computing power. Background Technology

[0002] During the operation of a data processing system, some edge nodes have limited computing power. When an edge node completes some coarse processing of the data and then needs to perform fine processing, it may experience computing power shortages. In this case, the node often needs to rent computing resources from other nodes to alleviate its own operational pressure and ensure the normal operation of the service.

[0003] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] This disclosure provides a method, apparatus, equipment, medium, and data processing system for renting computing power, which at least to some extent solves the problem of limited computing power at some edge nodes in related technologies, and the resulting computing power shortage when performing fine data processing.

[0005] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part from practice of this disclosure.

[0006] According to one aspect of this disclosure, a computing power rental method is provided, applied to a central scheduling system, the method comprising:

[0007] Receive computing power rental requests from edge nodes. The computing power rental request includes the address of the edge node, information on the computing power resources to be rented, rental time information, and rental fee threshold.

[0008] Based on the preset network topology, the addresses of edge nodes, and the information on rented computing resources, the rental cost for each node is calculated.

[0009] Based on the rental cost and rental cost threshold, multiple candidate nodes are determined;

[0010] Calculate the similarity value between the data used by candidate nodes and the data required by edge nodes within a preset time period;

[0011] Based on the rental cost and similarity value of each candidate node, the target rental node is determined from the candidate nodes.

[0012] In one embodiment of this disclosure, the method further includes:

[0013] The address and license certificate of the target lease node are sent to the edge node so that the edge node can establish a computing power lease relationship with the target lease node based on the address and license certificate of the target lease node.

[0014] In one embodiment of this disclosure, the license specification includes the following information:

[0015] Information on rented resources, addresses of edge nodes, and a signature from the central scheduling system confirming the legality of this computing power rental.

[0016] In one embodiment of this disclosure, vertices in the network topology graph are used to represent edge nodes, and edges in the network topology graph are used to represent the connection status between edge nodes. The weight of each edge is calculated by weighting the actual physical distance between edge nodes with the number of hops in the network routing link.

[0017] In one embodiment of this disclosure, the weight calculation formula for the edges in the network topology graph is as follows:

[0018] weight i =W d ·distance i +W h ·hop i

[0019] Where, weight i The distance is the weight of the edge from the endpoint node i to the starting node. i For actual physical distance, hop i W represents the hop count of a network routing link. d W represents the weights corresponding to the actual physical distance parameters. h The weight corresponding to the hop count of a network routing link.

[0020] In one embodiment of this disclosure, the rental fee for each node is calculated based on a preset network topology, the addresses of edge nodes, and the rented computing resources, including:

[0021] Based on the preset network topology and the addresses of the edge nodes, calculate the distance from each node in the network topology to the edge node that sent the resource expansion request. The distance value corresponding to each node is the sum of the weights of each edge on the path from the node to the edge node.

[0022] The rental cost of nodes is calculated in order of distance from smallest to largest, and the calculation stops when the rental cost of a node is greater than or equal to the rental cost threshold.

[0023] Based on rental costs and rental cost thresholds, multiple candidate nodes are identified, including:

[0024] Nodes whose rental cost is less than the rental cost threshold are selected as candidate nodes, thus obtaining a candidate node set.

[0025] In one embodiment of this disclosure, the method further includes:

[0026] In the candidate node set, delete nodes whose idle resources do not match the rental computing power resource information.

[0027] In one embodiment of this disclosure, the method further includes:

[0028] The activity level of each candidate node is calculated based on the total traffic of each candidate node within a preset period.

[0029] Based on the rental cost and similarity value of each candidate node, the target rental node is determined from the candidate nodes, including:

[0030] The target rental node is determined from the set of candidate nodes based on the activity level of each candidate node, the rental cost of each candidate node, and the similarity value.

[0031] According to another aspect of this disclosure, a computing power rental method is provided, applied to edge nodes, the method comprising:

[0032] A computing power rental request is sent to the central scheduling system. The request includes the address of the edge node, the information on the rented computing power resources, the rental time information, and the rental fee threshold. This allows the central scheduling system to calculate the rental fee for each node based on the preset network topology, the address of the edge node, and the information on the rented computing power resources. Based on the rental fee and the rental fee threshold, multiple candidate nodes are determined. The system also calculates the similarity value between the data used by the candidate nodes and the data required by the edge nodes within a preset time period. Finally, based on the rental fee and similarity value of each candidate node, the target rental node is determined from among the candidate nodes.

[0033] In one embodiment of this disclosure, the method further includes:

[0034] The target leased node's address and license certificate are sent by the central scheduling system.

[0035] Based on the address and license certificate of the target lease node, a computing power lease relationship is established with the target lease node.

[0036] In one embodiment of this disclosure, the license specification includes the following information:

[0037] Information on rented resources, addresses of edge nodes, and a signature from the central scheduling system confirming the legality of this computing power rental.

[0038] In one embodiment of this disclosure, establishing a computing power rental relationship with the target rental node based on the target rental node's address and license certificate includes:

[0039] The local scheduling system of the edge node sends the address of the target leased node, the address of the edge node, and the lease time information to the SDWAN orchestrator and controller, so that the SDWAN orchestrator and controller can establish an SDWAN channel between the target leased node and the edge node before the time corresponding to the lease time information.

[0040] During the time period corresponding to the lease time information, edge nodes transmit data and perform leased calculations through the SDWAN channel.

[0041] In one embodiment of this disclosure, the method further includes:

[0042] A computing power rental request is sent to the scheduling system of the target rental node, so that the target rental node can verify the address of the edge node that sent the computing power rental request according to the address of the edge node recorded in the license certificate, and return a computing power rental confirmation message after the verification is successful and the resources to be rented are selected.

[0043] In one embodiment of this disclosure, the method further includes:

[0044] Based on a preset computing power demand algorithm, determine whether edge nodes need to rent computing power;

[0045] Send a computing power rental request to the central scheduling system, including:

[0046] When edge nodes need to rent computing power, they send a computing power rental request to the central scheduling system.

[0047] In one embodiment of this disclosure, determining whether an edge node needs to lease computing power based on a preset computing power demand algorithm includes:

[0048] If, within a preset period, the resource utilization rate of resources in an edge node exceeds a preset utilization rate threshold for a period of time exceeding a preset time threshold, it is determined that the edge node needs to lease computing power.

[0049] In one embodiment of this disclosure, the method further includes:

[0050] When edge nodes need to rent computing power, an alert is sent to the data owner.

[0051] In one embodiment of this disclosure, the method further includes:

[0052] The threshold for rental fees sent by the receiving data owner.

[0053] According to another aspect of this disclosure, a computing power rental device is provided for use in a central scheduling system, the device comprising:

[0054] The rental request receiving module receives computing power rental requests from edge nodes. The computing power rental request includes the address of the edge node, information on the rented computing power resources, rental time information, and rental fee threshold.

[0055] The cost calculation module is used to calculate the rental cost for each node based on the preset network topology, the address of the edge node, and the rented computing resources.

[0056] The alternative node determination module determines multiple alternative nodes based on the rental fee and the rental fee threshold.

[0057] The similarity calculation module is used to calculate the similarity value between the data used by candidate nodes and the data required by edge nodes within a preset time period.

[0058] The rental node determination module is used to determine the target rental node from the candidate nodes based on the rental cost and similarity value of each candidate node.

[0059] According to another aspect of this disclosure, a computing power rental device is provided for use at an edge node, the device comprising:

[0060] The rental request sending module sends a computing power rental request to the central scheduling system. The computing power rental request includes the address of the edge node, the information of the rented computing power resources, the rental time information, and the rental fee threshold. This allows the central scheduling system to calculate the rental fee for each node based on the preset network topology, the address of the edge node, and the information of the rented computing power resources. Based on the rental fee and the rental fee threshold, multiple candidate nodes are determined. The system also calculates the similarity value between the data used by the candidate nodes and the data required by the edge nodes within a preset time. Finally, based on the rental fee and similarity value of each candidate node, the target rental node is determined from the candidate nodes.

[0061] According to another aspect of this disclosure, a data processing system is provided, the system comprising:

[0062] The edge layer consists of multiple edge nodes. The edge nodes adopt an integrated storage and computing architecture. When computing power needs to be leased, the edge nodes send computing power leasing requests to the central scheduling system. The computing power leasing request includes the address of the edge node, information on the leased computing power resources, leasing time information, and leasing fee threshold.

[0063] The central layer is equipped with a central scheduling system. The central scheduling system is used to receive computing power rental requests and calculate the rental fee for each node based on the preset network topology, the addresses of edge nodes, and the rental computing power resource information. Based on the rental fee and rental fee threshold, multiple candidate nodes are determined. The similarity value between the data used by the candidate nodes and the data required by the edge nodes is calculated within a preset time. Based on the rental fee and similarity value of each candidate node, the target rental node is determined from the candidate nodes.

[0064] According to another aspect of this disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to perform the above-described computing power rental method by executing the executable instructions.

[0065] According to another aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the above-described computing power rental method.

[0066] The computing power leasing method, apparatus, device, medium, and data processing system provided in this disclosure allow edge nodes with limited resources to send computing power leasing requests to a central scheduling system when these nodes are unable to handle the computing demands. The central scheduling system calculates the leasing fee for each node based on a preset network topology, the edge node's address, and the leased computing power resources. Based on the leasing fee and a threshold value, multiple candidate nodes are identified. The similarity value between the data used by the candidate nodes and the data requested by the edge nodes is calculated within a preset time period. Finally, based on the leasing fee and similarity value of each candidate node, a target leasing node is selected from the candidate nodes. In this way, edge nodes can lease the computing power resources of the target leasing node, thereby mitigating the impact of limited computing power on the overall operation of the data processing system.

[0067] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0068] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0069] Figure 1 This diagram illustrates the architecture of a data processing system according to an embodiment of the present disclosure.

[0070] Figure 2This diagram illustrates the architecture of another data processing system according to an embodiment of the present disclosure.

[0071] Figure 3 This diagram illustrates the architecture of yet another data processing system according to an embodiment of the present disclosure.

[0072] Figure 4 This diagram illustrates a flowchart of a computing power rental method according to an embodiment of the present disclosure.

[0073] Figure 5 This diagram illustrates another computing power rental method in an embodiment of the present disclosure.

[0074] Figure 6 This illustration shows a flowchart of yet another computing power rental method in an embodiment of this disclosure;

[0075] Figure 7 This diagram illustrates a flowchart of yet another computing power rental method according to an embodiment of the present disclosure;

[0076] Figure 8 This diagram illustrates a computing power rental device according to an embodiment of the present disclosure.

[0077] Figure 9 This diagram illustrates another computing power rental device in an embodiment of the present disclosure.

[0078] Figure 10 A structural block diagram of an electronic device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0079] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0080] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0081] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0082] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0083] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0084] As discussed in the background section, during the operation of a data processing system, some edge nodes have limited computing power. When an edge node completes some coarse processing of the data and then needs to perform fine processing, it may experience computing power shortages. In this case, the node often needs to rent computing resources from other nodes to alleviate its own operational pressure and ensure the normal operation of the service.

[0085] Figure 1 This diagram illustrates a data processing system provided in this disclosure, such as... Figure 1 As shown, the data processing system 100 provided in this embodiment includes a central layer 120 and an edge layer 140.

[0086] The edge layer 140 includes multiple edge nodes 141; the edge nodes 141 adopt an integrated storage and computing architecture; the edge nodes 141 are used to send computing power rental requests to the central scheduling system when computing power rental is required. The computing power rental request includes the address of the edge node, information on the rented computing power resources, rental time information, and rental fee threshold.

[0087] The central layer 120 is equipped with a central scheduling system. The central scheduling system is used to receive computing power rental requests and calculate the rental fee for each node based on the preset network topology, the addresses of edge nodes, and the rental computing power resource information. Based on the rental fee and rental fee threshold, multiple candidate nodes are determined. The similarity value between the data used by the candidate nodes and the data required by the edge nodes is calculated within a preset time. Based on the rental fee and similarity value of each candidate node, the target rental node is determined from the candidate nodes.

[0088] In some embodiments, the data processing system 100 may further include a data acquisition layer 160.

[0089] The data acquisition layer 160 is used to collect raw data at preset acquisition points 161 according to the needs of each business side, and send the raw data to the edge layer; each edge node 141 is used to store the raw data and send the target data in the raw data to the central layer 120; the central layer 120 is used to store the target data from the edge node 141, and in response to data processing instructions, schedule the edge nodes 141 of the edge layer 140 to cooperate with the central layer 120 to process the data.

[0090] The above parts are explained in detail below:

[0091] In some embodiments, such as Figure 2 As shown, the central layer 120 may include a central node 121; the central node 121 and the edge node 141 are connected via SD-WAN.

[0092] The number of central nodes 121 can be one or more, and is not limited here.

[0093] Big data is a core application and demand scenario for cloud-network convergence. However, the use of big data places higher demands on network and cloud collaboration, requiring flexible and dynamic adjustments to the deployment relationship between the cloud and the network. Therefore, big data requires cloud-network convergence, and a key driver of cloud-network convergence is the industrialization of big data and the empowerment of big data.

[0094] SD-WAN Network under the Integration of Data, Computing, Cloud, and Network: This integration provides deterministic and reliable connections between edge nodes and between central nodes and edge nodes, enabling secure, agile, and high-quality data transmission. It represents a new cloud-network convergence architecture. SD-WAN offers flexible networking, connecting central nodes, edge nodes, and between central and edge nodes to achieve secure, agile, and high-quality data transmission, facilitating unified management of data associated with different physical locations. In terms of security, encryption enhances security. Through unified orchestration, business language is translated into network requirements and security specifications based on data service / application characteristics. Regarding the security engine, security commands are generated to perform security hardening on links and network elements, such as tunnel encryption and traffic scrubbing. Network commands are generated to configure and adjust various physical / logical links, thereby enabling network control. This results in a "four-layer" + "one-control" + "one-code" architecture in network relationships.

[0095] This embodiment of the disclosure utilizes SDWAN control orchestration to leverage network advantages, match computing resources and data resources, and solve the problems of mismatched computing resources and data transmission bottlenecks in centralized architectures.

[0096] In some embodiments, the edge layer 140 is used to store, compute, and process local data according to local needs, while also undertaking off-site computing tasks from the central layer.

[0097] Edge nodes employ an integrated storage and computing architecture. As a core component of the integrated data and computing network architecture, they combine local network, computing, storage, and application capabilities to perform localized, specific business processing. The integrated storage and computing design of edge nodes facilitates data processing locally, significantly reducing the risk of data leakage and network load.

[0098] An edge node can consist of several parts, including a local data engine, edge storage resources, a computing power resource platform, and a resource scheduling platform.

[0099] The local data engine possesses the capabilities to cleanse, encrypt, interact with, and analyze data to meet local needs, enabling local data analysis and external empowerment. On the other hand, serving the central layer, for cross-provincial data analysis requests, edge nodes in various regions need to simplify and refine local data to form intermediate data that is then aggregated to the central layer for intelligent data fusion analysis.

[0100] Edge storage resources include data storage, preserving raw data after collection and cleaning; intermediate and final results of data analysis; and a repository of algorithm models for training. In this architecture, data is stored in edge cloud storage as much as possible for easy local allocation and use, minimizing data travel and ensuring computing power is as close to the edge as possible, reducing the load on centralized big data storage centers.

[0101] The computing resource platform enables local computation of data, including edge data intelligence fusion and edge federated learning. At the same time, for data that requires central-layer privacy computation, a privacy computation TEE channel is configured to communicate with the central node privacy computation platform.

[0102] The resource scheduling platform has two responsibilities: on the one hand, it undertakes the resource scheduling needs of the central layer resource scheduling center and provides the necessary storage and computing power and other joint scheduling services for the central node and other edge nodes; on the other hand, it is responsible for the resource scheduling of local data processing and analysis.

[0103] In this embodiment of the disclosure, the local data engine and computing resource platform, together with AI model algorithms, analyze and process local needs, and achieve computing power that moves with data through distributed edge computing deployment.

[0104] Edge resource scheduling centers handle local data analysis and processing, and work in conjunction with central resource scheduling to optimize resources across the entire network, effectively reducing latency and improving performance.

[0105] The aforementioned data processing system changes the traditional centralized cloud resource model. By using technologies such as distributed computing, resource migration, and virtualization, it fully utilizes the resource capabilities of the cloud and edge, optimizes resource allocation, and improves resource utilization.

[0106] Among them, the main entities are connected through an SD-WAN network, the entire network achieves global resource scheduling through a global dispatch center, and data assets are managed in a refined manner and data lineage is traced through data identification.

[0107] During the normal operation of the aforementioned data processing system, when an edge node malfunctions due to insufficient computing or storage resources, causing the system to fail, there will be a need to expand the edge node's capacity. In this case, the edge node needs to migrate some of its data to other edge nodes or data center nodes for storage to alleviate its own operational pressure and ensure normal service operation.

[0108] The following detailed description of this exemplary implementation method is provided in conjunction with the accompanying drawings and embodiments.

[0109] Figure 3 This diagram illustrates a data processing system structure according to an embodiment of the present disclosure, such as... Figure 3 As shown, the data processing system 300 may include an SD-WAN controller 301. The SD-WAN controller 301 can be used to establish SD-WAN connections between nodes in response to control commands from the central scheduling system 302.

[0110] SDWAN controllers, software-defined wide area network technology, based on the Internet and 4G / 5G, enable enterprise customers to quickly build a private enterprise cloud network isolated from the public Internet at a lower cost, and provide multi-point on-demand interconnection services between sites and the cloud, helping IT systems and networks to quickly form networks.

[0111] In some embodiments, such as Figure 3 As shown, the data processing system may also include a cloud management system, which is not shown in the figure. The cloud management system can be used to coordinate with the central scheduling system to schedule the expansion of storage and computing resources when there is a request for cloud resource expansion.

[0112] In response to the uncertainties of data and algorithms in this network architecture, the cloud management system should also be demand-driven, coordinately schedule data and computing resources, and achieve "data + algorithm" collaborative scheduling through cloud-edge collaboration.

[0113] In some embodiments, such as Figure 3 As shown, the data processing system may also include a data middle platform 303, which can be used to send the data address required by the remote node to the remote node when there is a data demand from the remote node in the network.

[0114] Data Platform 303 empowers internal and external data applications through its DaaS service engine, while also enabling data asset management through data weaving technology.

[0115] The central scheduling system 302 in the above embodiment is located in the central layer. The central layer may include a central storage node, a computing power center, a privacy computing center, a data sharing and exchange center, a data intelligence fusion center, a federated learning center, a data middle platform, and a resource scheduling center. In terms of location, it can form a regional configuration of "data in the east and computing in the west".

[0116] Central storage: Compared to the edge side, central storage nodes have a larger storage capacity and mainly store core network data, non-core data of the edge layer, intermediate edge data that needs to be calculated and processed by the central layer, and central layer analysis results data, etc.

[0117] Computing Power Center: The computing power center is loosely coupled with the central layer applications. It mainly undertakes general computing tasks for both the central layer and the edge layer. When the local computing power of the edge layer is insufficient, it can rent the computing power center of the central layer for massive computing. The central layer applications can call the computing power resources of the computing power center to execute computing tasks.

[0118] Data Platform: The data platform empowers internal and external data applications through the DaaS service engine, and at the same time, realizes data asset management based on data weaving technology.

[0119] Resource Scheduling Center: The resource scheduling center is the central brain of the entire integrated data, computing, cloud, and network architecture. It is responsible for scheduling the network's storage resources, computing resources, SD-WAN network resources, and data resources. All network service requests are first allocated to matching capability centers based on resource usage through the global scheduling center. Simultaneously, considering the uncertainty of data and algorithms in this network architecture, demand-driven, collaborative scheduling of data and computing resources, along with cloud-edge collaboration, should be implemented to achieve collaborative scheduling of "data + algorithm".

[0120] In this embodiment of the disclosure, the central layer may include several functional modules such as central storage, computing power center, and resource scheduling center. Each module is connected to the cloud and edge through the brain of the resource scheduling center. Different modules are scheduled to different centers for execution according to different needs, so as to realize demand-driven algorithm, algorithm-driven data, data-driven collection, data matching resources according to algorithm, optimized distribution, and stimulate the enthusiasm of all parties involved through data value, and build an internal and external data ecosystem.

[0121] Figure 4 This diagram illustrates a resource expansion method according to an embodiment of the present disclosure, such as... Figure 4 As shown, the method includes the following steps:

[0122] S402, the edge node sends a computing power rental request to the central scheduling system. The computing power rental request includes the edge node's address, information on the rented computing power resources, rental time information, and rental fee threshold.

[0123] S404, the central scheduling system calculates the rental cost for each node based on the preset network topology, the addresses of edge nodes, and the rental computing resources.

[0124] S406, the central dispatch system determines multiple alternative nodes based on rental fees and rental fee thresholds;

[0125] S408, The central scheduling system calculates the similarity value between the data used by the candidate nodes and the demand data of the edge nodes within a preset time.

[0126] S410, the central scheduling system determines the target rental node from the candidate nodes based on the rental cost and similarity value of each candidate node.

[0127] The above steps are explained in detail below:

[0128] It should be noted that after S410, the central scheduling system can also send the address and license certificate of the target lease node to the edge node, so that the edge node can establish a computing power lease relationship with the target lease node based on the address and license certificate of the target lease node.

[0129] The license certificate includes the following information:

[0130] Information on rented resources, addresses of edge nodes, and a signature from the central scheduling system confirming the legality of this computing power rental.

[0131] In some embodiments, establishing a computing power leasing relationship with the target lease node based on the target lease node's address and license certificate includes: sending the target lease node's address, the edge node's address, and lease time information to the SDWAN orchestrator and controller through the edge node's local scheduling system, so that the SDWAN orchestrator and controller establish an SDWAN channel between the target lease node and the edge node before the time corresponding to the lease time information; during the time corresponding to the lease time information, the edge node performs data transmission and leased computation through the SDWAN channel.

[0132] As an example, the local scheduling system of the edge node where the data resides sends the address information, lease period information, and required bandwidth information of the edge node and the computing power leased node to the SDWAN orchestrator and controller.

[0133] The SDWAN orchestrator and controller establish SDWAN channels between edge nodes and leased computing power nodes before the computing power lease begins. After the channels are established, the edge nodes transmit data and perform leased computations at the start of the lease period.

[0134] The central scheduling system is responsible for statistics and monitoring of the data usage of each edge node in the data computing cloud network. When the computing power resources of the edge nodes are insufficient, it coordinates with the scheduling systems of each edge node to complete the task of renting computing power.

[0135] There are two ways for edge nodes to discover computing power rental needs: nodes actively requesting and computing resource alerts.

[0136] Node-initiated request method: Before an edge node performs a computing task, it conducts a computing power risk assessment based on the edge node's hardware, historical computing tasks, and other conditions. Based on the assessment results, it analyzes whether the edge node has a computing power rental requirement. If so, it needs to send a computing power rental request to the central scheduling system.

[0137] Computing resource early warning method: This method is mainly applicable to situations where edge nodes have sudden computing power demands in computing tasks.

[0138] Once the data processing system architecture is successfully established, the edge nodes set the computing power requirement algorithm according to their own characteristics.

[0139] As an example, the preset computing power requirement algorithm can be as follows:

[0140] Set n, TH, Rt j The parameters are used to continuously determine whether the edge node has a computing power rental demand H(x) with a period of TH.

[0141] Where n represents the n types of computing resources available at the edge node, including CPU resources, GPU resources, and other special computing resources.

[0142] Rt j Let be the threshold for the utilization rate of the j-th resource.

[0143]

[0144] U(·) is the step function:

[0145]

[0146] R j Let be the utilization rate of the j-th type of computing power resource.

[0147] T j Let R be the utilization rate of the j-th type of computing resource within the time period TH. j Exceeding threshold Rt j The total usage time. That is, every time an edge node determines whether it has a computing power rental demand H(x) within the TH time interval, if there is a j-th type of computing power resource utilization rate R. j Exceeding threshold Rt jIn cases where the resource type exceeds the threshold, record the usage time exceeding the threshold and record the total of this time as T. j .

[0148] TV j Preset the usage time threshold for the j-th type of computing power resource for edge nodes.

[0149] If within the TH time interval, T j Beyond TV j If so, it is determined that the edge node has a demand for computing power rental.

[0150] When H(x) > 0, it indicates that the edge node is experiencing a shortage of computing resources in the near future, and the edge node sends an alert to the data owner (the data owner can be regarded as an entity or application).

[0151] Accordingly, the method disclosed herein may also include the following steps:

[0152] Based on a preset computing power demand algorithm, determine whether edge nodes need to rent computing power;

[0153] Send a computing power rental request to the central scheduling system, including:

[0154] When edge nodes need to rent computing power, they send a computing power rental request to the central scheduling system.

[0155] In some embodiments, based on a preset computing power demand algorithm, determining whether an edge node needs to lease computing power includes:

[0156] If, within a preset period, the resource utilization rate of resources in an edge node exceeds a preset utilization rate threshold for a period of time exceeding a preset time threshold, it is determined that the edge node needs to lease computing power.

[0157] In some embodiments, when edge nodes need to rent computing power, an alert can be sent to the data owner.

[0158] In the example above, when H(x) > 0, the edge node confirms the discovery of a computing power rental requirement and needs to send a computing power rental request to the central scheduling system.

[0159] As an example, edge nodes that need to rent computing power can divide computing tasks into local computing tasks and rented computing tasks, and generate the parameters required in the rental request.

[0160] A computing power rental request includes parameters such as the address of the edge node, information on the rented computing resources, rental time information, and rental fee threshold.

[0161] As an example, a computing power rental request includes: the location of the edge node (such as IP and MAC address), the type and size of the computing power resources to be rented, the type and size of the data to be rented, the acceptable rental period for the edge node, and the maximum fee that can be paid for the computing power rental.

[0162] The location of an edge node can be represented by parameters such as IP and MAC address.

[0163] The types and sizes of rented computing resources are determined by R in step 1. j The parameters determine the required rental resource size Re for j types of computing power resources. j .

[0164] The data type and size to be rented are estimated based on the computational task, resulting in the data set D that needs to be transmitted to other edge nodes for computation. k (k = 1, 2, 3), where D k These represent the sizes of structured data, semi-structured data, and unstructured data, respectively.

[0165] The acceptable computing power rental period for this edge node is calculated based on the node's historical resource utilization rate. The time with the lowest historical resource utilization rate is selected for computing power rental to reduce the impact of the computing power rental process on the normal operation of the node.

[0166] Let C_max be the maximum fee that can be paid for renting computing power, which is set by the user owner.

[0167] Accordingly, the above method may also include: the edge node receiving the rental fee threshold sent by the data owner.

[0168] The central scheduling system determines the optimal computing power rental target based on the information uploaded by the edge nodes requesting computing power rental, combined with the resource usage of the central node and other edge nodes, and the rental node selection strategy.

[0169] When building a data processing system, the central scheduling system can maintain a network topology map that records all nodes in the cloud network, which is the network topology map in S404 mentioned above.

[0170] In the network topology graph, vertices represent edge nodes, and edges represent the connection status between edge nodes. The weight of each edge is calculated by weighting the actual physical distance between edge nodes with the number of hops in the network routing link.

[0171] As an example, the formula for calculating the edge weights in a network topology graph is as follows:

[0172] weight i =W d·distance i +W h ·hop i

[0173] Where, weight i The distance is the weight of the edge from the endpoint node i to the starting node. i For actual physical distance, hop i W represents the hop count of a network routing link. d W represents the weights corresponding to the actual physical distance parameters. h The weight corresponding to the hop count of a network routing link.

[0174] In some embodiments, the central scheduling system calculates the distance from each node to the requesting node using Dijkstra's algorithm, according to the network topology of the data processing system. (The distance is the sum of the weights on the path formed by the two nodes, with the requesting node as the starting point and the target lease node as the ending point.)

[0175] Based on the type and size of resources that the edge node sending the request needs to rent, calculate the rental cost C of other edge nodes in ascending order of distance. i until C is satisfied i If the value is greater than or equal to C_max, then the nodes that have been traversed are extracted as a set of candidate nodes.

[0176] The remaining idle computing resources (Re_free) of the candidate leased nodes in the computation set. i-j and free storage resources S_free i Record Re_free i-j S_free i Nodes that do not meet the requested resource size are removed from the set of candidate lease nodes.

[0177] The central scheduling system calculates the activity level of each candidate node, based on the total traffic flow over the past three months. i To decide, that is:

[0178] Flow i =Flow_up i +Flow_down i

[0179] Among them, Flow_up i Flow_down represents the node's uplink traffic over the past three months. i This represents the node's downlink traffic over the past three months. Flow i The higher the value, the more suitable the node is as a leased node.

[0180] The central scheduling system calculates the similarity Si of each candidate node. i This metric is determined by the similarity between the usage data of candidate nodes and the demand data of requesting nodes over the past three months. i The higher the value, the more suitable the edge node is for handling the computational tasks of the requesting node.

[0181] The rental cost C of the central dispatch system for the integrated edge nodes i Node activity flow i Node similarity Si i This determines the optimal lease node, which is the target lease node.

[0182] The central scheduling system returns the target node's network address and a license certificate authorizing the rental of computing power to the edge node that sent the request, and notifies the target node to prepare to receive data.

[0183] The computing power rental license includes the type and size of the rented resources, the physical address of the edge node that sent the request, and the signature of the central scheduling system confirming the legality of the computing power rental.

[0184] In some embodiments, the method may further include: sending a computing power rental request to the scheduling system of the target rental node, so that the target rental node verifies the address of the edge node sending the computing power rental request according to the address of the edge node recorded in the license certificate, and returns a computing power rental confirmation message after the verification is successful and the resources for rental are selected.

[0185] As an example, the process of sending a computing power rental request to the target rental node can be as follows:

[0186] 1) After obtaining the address of the computing power rental node from the central scheduling system, the requesting node sends a computing power rental request to the scheduling system of the rental node.

[0187] 2) The computing power rental node compares the physical address of the requesting node with the rental data license certificate sent by the central scheduling system to see if they are consistent.

[0188] 3) The computing power rental node selects the optimal server or VM address information and returns it to the data owner based on the specific rental requirements and the utilization rate of each local server or VM.

[0189] 4) Once authentication and data reception preparation are complete, a confirmation message will be returned to the requesting node.

[0190] The node requested above is the edge node introduced earlier.

[0191] Compared to computing power rental schemes in related technologies, the computing power rental behavior between edge nodes in this disclosure relies on the coordination of a central scheduling system. The central scheduling system stores and maintains a network topology diagram of the edge nodes in the data computing cloud network, and determines the target node for computing power rental based on an optimal target rental node algorithm. The type and size of the resources to be rented, the overall rental cost, and the data transmission time are all calculated using relevant algorithms to achieve an efficient, secure, and complete computing power rental process.

[0192] Based on the same inventive concept, this disclosure also provides a method for renting computing power, such as... Figure 5 As shown, the method includes the following steps:

[0193] S501 accepts and inputs the relevant parameters for renting computing power to the requesting node.

[0194] Here, the requesting node is the edge node mentioned earlier. For details on computing power rental parameters, please refer to the previous section on computing power rental requests; they will not be repeated here.

[0195] S502 accepts and inputs the actual physical distance between edge nodes and the number of network routing hops.

[0196] S503 calculates the link weights between nodes and generates a network topology diagram.

[0197] Here, S502 and S503 can be steps that are completed in advance.

[0198] S504, Dijkstra calculates the distance from each node to the requesting node and calculates the rental cost based on the resource type, size, and distance.

[0199] S505 determines whether the rental cost exceeds the maximum cost that the requesting node can pay.

[0200] If the limit is exceeded, execute S506 to add the node to the set of alternative lease nodes and return S504.

[0201] If the maximum cost is not exceeded, execute S507 to obtain a set of alternative leased nodes and sort them by cost.

[0202] The alternative rental nodes here are the alternative nodes mentioned above.

[0203] S508 determines whether the available resources of the candidate node are greater than the required rental resources.

[0204] If the value is not greater than the specified value, execute S509 to delete the candidate node.

[0205] If the activity level is greater than the target level, execute S510 to calculate and sort the activity levels of the candidate nodes.

[0206] S511, determine if the activity level is less than the set threshold.

[0207] If the value is not less than the specified value, execute S512 to delete the candidate node.

[0208] If the similarity is less than 1, execute S513 to calculate and sort the candidate nodes.

[0209] S514, determine whether the similarity is greater than the set threshold.

[0210] If the value is not greater than the specified value, execute S515 to delete the candidate node.

[0211] If the value is greater than the target node, execute S516 to send the target node's network address and computing power rental license certificate to the requesting node.

[0212] In this embodiment of the disclosure, the target leased node is selected by calculation through the central scheduling system. In complex network and resource environments, the optimal leased node can be selected based on different computing power leasing needs and the network status and node resource status of edge nodes, so as to realize intelligent control of the computing power of the entire network and free allocation of computing tasks.

[0213] Based on the same inventive concept, this disclosure also provides a computing power rental method, applied to a central scheduling system, such as... Figure 6 As shown, the method includes:

[0214] S602, receive a computing power rental request from an edge node. The computing power rental request includes the address of the edge node, information on the rented computing power resources, rental time information, and rental fee threshold.

[0215] S604 calculates the rental cost for each node based on the preset network topology, the addresses of edge nodes, and the rental computing resources.

[0216] S606 determines multiple candidate nodes based on rental fees and rental fee thresholds;

[0217] S608, calculate the similarity value between the data used by candidate nodes and the data required by edge nodes within a preset time period;

[0218] S610, determine the target rental node from the candidate nodes based on the rental cost and similarity value of each candidate node.

[0219] In some embodiments, the method may further include:

[0220] The address and license certificate of the target lease node are sent to the edge node so that the edge node can establish a computing power lease relationship with the target lease node based on the address and license certificate of the target lease node.

[0221] In some embodiments, the license specification includes the following information:

[0222] Information on rented resources, addresses of edge nodes, and a signature from the central scheduling system confirming the legality of this computing power rental.

[0223] In some embodiments, vertices in the network topology graph represent edge nodes, and edges in the network topology graph represent the connection status between edge nodes. The weight of each edge is calculated by weighting the actual physical distance between edge nodes with the number of hops in the network routing link.

[0224] In some embodiments, the weights of edges in a network topology graph are calculated using the following formula:

[0225] weight i =W d ·distance i +W h ·hop i

[0226] Where, weight i The distance is the weight of the edge from the endpoint node i to the starting node. i For actual physical distance, hop i W represents the hop count of a network routing link. d W represents the weights corresponding to the actual physical distance parameters. h The weight corresponding to the hop count of a network routing link.

[0227] In some embodiments, S604 is specifically used for:

[0228] Based on the preset network topology and the addresses of the edge nodes, calculate the distance from each node in the network topology to the edge node that sent the resource expansion request. The distance value corresponding to each node is the sum of the weights of each edge on the path from the node to the edge node.

[0229] The rental cost of nodes is calculated in descending order of distance value until the rental cost of a node is greater than or equal to the rental cost threshold.

[0230] Correspondingly, S604 can be a set of candidate nodes, where the rental cost is less than the rental cost threshold.

[0231] In some embodiments, the method may further include:

[0232] In the candidate node set, delete nodes whose idle resources do not match the rental computing power resource information.

[0233] In some embodiments, the method may further include:

[0234] The activity level of each candidate node is calculated based on the total traffic of each candidate node within a preset period.

[0235] Accordingly, S610 can be: determining the target rental node from the set of candidate nodes based on the activity level of each candidate node, the rental cost of each candidate node, and the similarity value.

[0236] Based on the same inventive concept, this disclosure also provides a computing power rental method applied to edge nodes, such as... Figure 7 As shown, the method includes:

[0237] S702, a computing power rental request is sent to the central scheduling system. The computing power rental request includes the address of the edge node, the information of the rented computing power resources, the rental time information, and the rental fee threshold. This enables the central scheduling system to calculate the rental fee corresponding to each node based on the preset network topology, the address of the edge node, and the information of the rented computing power resources. Based on the rental fee and the rental fee threshold, multiple candidate nodes are determined, and the similarity value between the data used by the candidate nodes and the data required by the edge nodes within a preset time period is calculated. Finally, based on the rental fee and similarity value of each candidate node, the target rental node is determined from the candidate nodes.

[0238] In some embodiments, the method may further include:

[0239] The target leased node's address and license certificate are sent by the central scheduling system.

[0240] Based on the address and license certificate of the target lease node, a computing power lease relationship is established with the target lease node.

[0241] In some embodiments, the license specification includes the following information:

[0242] Information on rented resources, addresses of edge nodes, and a signature from the central scheduling system confirming the legality of this computing power rental.

[0243] In some embodiments, establishing a computing power rental relationship with the target rental node based on the target rental node's address and license certificate includes:

[0244] The local scheduling system of the edge node sends the address of the target leased node, the address of the edge node, and the lease time information to the SDWAN orchestrator and controller, so that the SDWAN orchestrator and controller can establish an SDWAN channel between the target leased node and the edge node before the time corresponding to the lease time information.

[0245] During the time period corresponding to the lease time information, edge nodes transmit data and perform leased calculations through the SDWAN channel.

[0246] In some embodiments, the method may further include:

[0247] A computing power rental request is sent to the scheduling system of the target rental node, so that the target rental node can verify the address of the edge node that sent the computing power rental request according to the address of the edge node recorded in the license certificate, and return a computing power rental confirmation message after the verification is successful and the resources to be rented are selected.

[0248] In some embodiments, the method may further include:

[0249] Based on a preset computing power demand algorithm, it is determined whether edge nodes need to rent computing power.

[0250] In some embodiments, based on a preset computing power demand algorithm, determining whether an edge node needs to lease computing power includes:

[0251] If, within a preset period, the resource utilization rate of resources in an edge node exceeds a preset utilization rate threshold for a period of time exceeding a preset time threshold, it is determined that the edge node needs to lease computing power.

[0252] In some embodiments, the method may further include:

[0253] When edge nodes need to rent computing power, an alert is sent to the data owner.

[0254] In some embodiments, the method may further include:

[0255] The threshold for rental fees sent by the receiving data owner.

[0256] Based on the same inventive concept, this disclosure also provides a computing power rental device, as described in the following embodiments. Since the principle by which this device solves the problem is similar to that of the above-described method embodiments, the implementation of this device embodiment can refer to the implementation of the above-described method embodiments, and repeated details will not be elaborated further.

[0257] Figure 8 This invention discloses a computing power rental device in an embodiment of the present disclosure, applied to a central scheduling system, such as... Figure 8 As shown, the computing power rental device 800 includes:

[0258] The rental request receiving module 802 receives computing power rental requests from edge nodes. The computing power rental request includes the address of the edge node, information on the rented computing power resources, rental time information, and rental fee threshold.

[0259] The cost calculation module 804 is used to calculate the rental cost of each node based on the preset network topology, the address of the edge node, and the rented computing resources.

[0260] The alternative node determination module 806 determines multiple alternative nodes based on the rental fee and the rental fee threshold.

[0261] The similarity calculation module 808 is used to calculate the similarity value between the data used by candidate nodes and the data required by edge nodes within a preset time period.

[0262] The rental node determination module 810 is used to determine the target rental node from the candidate nodes based on the rental cost and similarity value of each candidate node.

[0263] In some embodiments, the computing power rental device 800 may further include:

[0264] The second information sending module is used to send the address and license certificate of the target lease node to the edge node, so that the edge node can establish a computing power lease relationship with the target lease node based on the address and license certificate of the target lease node.

[0265] In some embodiments, the license specification includes the following information:

[0266] Information on rented resources, addresses of edge nodes, and a signature from the central scheduling system confirming the legality of this computing power rental.

[0267] In some embodiments, vertices in the network topology graph represent edge nodes, and edges in the network topology graph represent the connection status between edge nodes. The weight of each edge is calculated by weighting the actual physical distance between edge nodes with the number of hops in the network routing link.

[0268] In some embodiments, the weights of edges in a network topology graph are calculated using the following formula:

[0269] weight i =W d ·distance i +W h ·hop i

[0270] Where, weight i The distance is the weight of the edge from the endpoint node i to the starting node. i For actual physical distance, hop i W represents the hop count of a network routing link. d W represents the weights corresponding to the actual physical distance parameters. h The weight corresponding to the hop count of a network routing link.

[0271] In some embodiments, the cost calculation module 804 can be implemented as follows:

[0272] Based on the preset network topology and the addresses of the edge nodes, calculate the distance from each node in the network topology to the edge node that sent the resource expansion request. The distance value corresponding to each node is the sum of the weights of each edge on the path from the node to the edge node.

[0273] The rental cost of nodes is calculated in descending order of distance value until the rental cost of a node is greater than or equal to the rental cost threshold.

[0274] Correspondingly, the candidate node determination module 806 can be used to select nodes whose rental fees are less than the rental fee threshold as candidate nodes, thereby obtaining a candidate node set.

[0275] In some embodiments, the computing power rental device 800 may further include:

[0276] The node deletion module is used to delete nodes from the candidate node set whose idle resources do not match the rental computing power resource information.

[0277] In some embodiments, the computing power rental device 800 may further include:

[0278] The activity calculation module is used to calculate the activity of each candidate node based on the total traffic of each candidate node within a preset period.

[0279] Correspondingly, the rental node determination module 810 can be used to determine the target rental node from the set of candidate nodes based on the activity level of each candidate node, the rental cost of each candidate node, and the similarity value.

[0280] The computing power rental device provided in this application embodiment can be used to execute the computing power rental methods provided in the above-described method embodiments. The implementation principle and technical effect are similar, and will not be described in detail here for the sake of brevity.

[0281] Based on the same inventive concept, this disclosure also provides a computing power rental device, applied to edge nodes, such as... Figure 9 As shown, the computing power rental device 900 includes:

[0282] The rental request sending module 902 sends a computing power rental request to the central scheduling system. The computing power rental request includes the address of the edge node, the information on the rented computing power resources, the rental time information, and the rental fee threshold. This enables the central scheduling system to calculate the rental fee corresponding to each node based on the preset network topology, the address of the edge node, and the information on the rented computing power resources. Based on the rental fee and the rental fee threshold, the system determines multiple candidate nodes and calculates the similarity value between the data used by the candidate nodes and the data required by the edge nodes within a preset time. Finally, based on the rental fee and similarity value of each candidate node, the system determines the target rental node from among the candidate nodes.

[0283] In some embodiments, the computing power rental device 900 may further include:

[0284] The second information receiving module is used to receive the address and license certificate of the target leased node sent by the central scheduling system;

[0285] The lease relationship construction module is used to establish a computing power lease relationship with the target lease node based on the address and license certificate of the target lease node.

[0286] In some embodiments, the license specification includes the following information:

[0287] Information on rented resources, addresses of edge nodes, and a signature from the central scheduling system confirming the legality of this computing power rental.

[0288] In some embodiments, the lease relationship construction module may be specifically used for:

[0289] The local scheduling system of the edge node sends the address of the target leased node, the address of the edge node, and the lease time information to the SDWAN orchestrator and controller, so that the SDWAN orchestrator and controller can establish an SDWAN channel between the target leased node and the edge node before the time corresponding to the lease time information.

[0290] During the time period corresponding to the lease time information, edge nodes transmit data and perform leased calculations through the SDWAN channel.

[0291] In some embodiments, the computing power rental device 900 may further include:

[0292] The rental request sending module is used to send a computing power rental request to the scheduling system of the target rental node, so that the target rental node can verify the address of the edge node sending the computing power rental request according to the address of the edge node recorded in the license certificate, and return a computing power rental confirmation message after the verification is successful and the resources to be rented are selected.

[0293] In some embodiments, the computing power rental device 900 may further include:

[0294] The demand judgment module is used to determine whether edge nodes need to rent computing power based on a preset computing power demand algorithm.

[0295] In some embodiments, the demand determination module is specifically used to determine that the edge node needs to rent computing power when the resource utilization rate of the resources in the edge node is greater than a preset utilization rate threshold for a period of time within a preset period.

[0296] In some embodiments, the computing power rental device 900 may further include:

[0297] The alarm sending module is used to send alarms to the data owner when edge nodes need to rent computing power.

[0298] In some embodiments, the computing power rental device 900 may further include:

[0299] The fee threshold receiving module is used to receive the rental fee threshold sent by the data owner.

[0300] The computing power rental device provided in this application embodiment can be used to execute the computing power rental methods provided in the above-described method embodiments. The implementation principle and technical effect are similar, and will not be described in detail here for the sake of brevity.

[0301] Those skilled in the art will understand that various aspects of this disclosure can be implemented as a system, method, or program product. Therefore, various aspects of this disclosure can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, collectively referred to herein as a "circuit," "module," or "system."

[0302] The following reference Figure 10 To describe an electronic device 1000 according to such an embodiment of the present disclosure. Figure 10 The electronic device 1000 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments disclosed herein.

[0303] like Figure 10 As shown, the electronic device 1000 is manifested in the form of a general-purpose computing device. The components of the electronic device 1000 may include, but are not limited to: at least one processing unit 1010, at least one storage unit 1020, and a bus 1030 connecting different system components (including storage unit 1020 and processing unit 1010).

[0304] The storage unit stores program code that can be executed by the processing unit 1010, causing the processing unit 1010 to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure. For example, the processing unit 1010 can perform the steps of the above-described method embodiments.

[0305] Storage unit 1020 may include readable media in the form of volatile storage units, such as random access memory (RAM) 10201 and / or cache memory 10202, and may further include read-only memory (ROM) 10203.

[0306] Storage unit 1020 may also include a program / utility 10204 having a set (at least one) program module 10205, such program module 10205 including but not limited to: operating system, one or more application programs, other program modules and program data, each or some combination of these examples may include an implementation of a network environment.

[0307] Bus 1030 can represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local bus using any of the multiple bus structures.

[0308] The electronic device 1000 can also communicate with one or more external devices 1040 (e.g., keyboard, pointing device, Bluetooth device, etc.), and with one or more devices that enable a user to interact with the electronic device 1000, and / or with any device that enables the electronic device 1000 to communicate with one or more other computing devices (e.g., router, modem, etc.). Such communication can be performed through the input / output (I / O) interface 1050.

[0309] Furthermore, the electronic device 1000 can also communicate with one or more networks (such as local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via the network adapter 1060.

[0310] like Figure 10 As shown, network adapter 1060 communicates with other modules of electronic device 1000 via bus 1030.

[0311] It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 1000, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0312] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, terminal device, or network device, etc.) to execute the methods according to the embodiments of this disclosure.

[0313] In exemplary embodiments of this disclosure, a computer-readable storage medium is also provided, which may be a readable signal medium or a readable storage medium. A program product capable of implementing the methods described above is stored thereon.

[0314] In some possible implementations, various aspects of this disclosure may also be implemented as a program product comprising program code that, when run on a terminal device, causes the terminal device to perform the steps described in the “Exemplary Methods” section of this specification according to various exemplary embodiments of this disclosure.

[0315] More specific examples of computer-readable storage media in this disclosure may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0316] In this disclosure, a computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, wherein readable program code is carried.

[0317] The transmitted data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0318] A readable signal medium can also be any readable medium other than a readable storage medium, which can send, propagate or transmit a program for use by or in connection with an instruction execution system, apparatus or device.

[0319] In some examples, program code contained on a computer-readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0320] In practice, program code for performing the operations of this disclosure can be written using any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages.

[0321] The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0322] In cases involving remote computing devices, the remote computing devices can be connected to user computing devices via any type of network, including local area networks (LANs) or wide area networks (WANs), or they can be connected to external computing devices (e.g., via the Internet using an Internet service provider).

[0323] It should be noted that although several modules or units of the device used for action execution are mentioned in the detailed description above, this division is not mandatory.

[0324] In fact, according to embodiments of this disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0325] Furthermore, although the steps of the method in this disclosure are described in a specific order in the accompanying drawings, this does not require or imply that the steps must be performed in that specific order, or that all the steps shown must be performed to achieve the desired result. Additional or alternative steps may be omitted, multiple steps may be combined into one step, and / or a step may be broken down into multiple steps.

[0326] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware.

[0327] Therefore, the technical solution according to the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, mobile hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, mobile terminal, or network device, etc.) to execute the method according to the embodiments of this disclosure.

[0328] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein.

[0329] This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the appended claims.

Claims

1. A method for renting computing power, characterized in that, Applied to a central scheduling system, the method includes: Receive computing power rental requests from edge nodes, the computing power rental requests including the address of the edge node, information on the rented computing power resources, rental time information, and rental fee threshold; Based on the preset network topology, the addresses of the edge nodes, and the rented computing resources, the rental cost for each node is calculated. Based on the rental fee and the rental fee threshold, multiple candidate nodes are determined; The activity level of each candidate node is calculated based on the total traffic of each candidate node within a preset period. Calculate the similarity value between the data used by the candidate nodes and the demand data of the edge nodes within a preset time period; Based on the activity level of each candidate node, the rental cost of each candidate node, and the similarity value, a target rental node is determined from the candidate nodes; wherein, the higher the activity level, the more suitable it is as the target rental node, and the higher the similarity value, the more suitable it is as the target rental node.

2. The method according to claim 1, characterized in that, The method further includes: The address and license certificate of the target lease node are sent to the edge node so that the edge node can establish a computing power lease relationship with the target lease node based on the address and license certificate of the target lease node.

3. The method according to claim 2, characterized in that, The license certificate includes the following information: Information on rented resources, the address of the edge node, and the signature of the central scheduling system confirming the legality of this computing power rental.

4. The method according to claim 1, characterized in that, The vertices in the network topology graph represent edge nodes, and the edges in the network topology graph represent the connection status between edge nodes. The weight of each edge is calculated by weighting the actual physical distance between edge nodes with the number of hops in the network routing link.

5. The method according to claim 4, characterized in that, The formula for calculating the edge weights in a network topology graph is as follows: in, Let be the weight of the edge from the endpoint node i to the starting node. This refers to the actual physical distance. This refers to the hop count of network routing links. The weights corresponding to the actual physical distance parameters. The weight corresponding to the hop count of a network routing link.

6. The method according to claim 4, characterized in that, Based on the preset network topology, the addresses of the edge nodes, and the rented computing resources, the rental fee for each node is calculated, including: Based on the preset network topology diagram and the address of the edge node, calculate the distance value from each node in the network topology diagram to the edge node that sent the computing power rental request. The distance value corresponding to each node is the sum of the weights of each edge on the path from the node to the edge node. The rental cost of a node is calculated in ascending order of distance values ​​until the rental cost of the node is greater than or equal to the rental cost threshold, at which point the calculation stops. The determination of multiple candidate nodes based on the rental fee and the rental fee threshold includes: Nodes whose rental fees are less than the rental fee threshold are selected as candidate nodes, thus obtaining a candidate node set.

7. The method according to claim 6, characterized in that, The method further includes: In the set of candidate nodes, nodes whose idle resources do not match the rented computing power resource information are deleted.

8. A method for renting computing power, characterized in that, Applied to edge nodes, the method includes: A computing power rental request is sent to the central scheduling system. The computing power rental request includes the address of the edge node, the information of the rented computing power resources, the rental time information, and the rental fee threshold. This allows the central scheduling system to calculate the rental fee for each node based on a preset network topology, the address of the edge node, and the information of the rented computing power resources. Based on the rental fee and the rental fee threshold, multiple candidate nodes are determined. The activity level of each candidate node is calculated based on the total traffic of each candidate node within a preset period. The similarity value between the data used by the candidate nodes and the demand data of the edge nodes within a preset period is calculated. Finally, based on the activity level of each candidate node, the rental fee of each candidate node, and the similarity value, a target rental node is determined from the candidate nodes. Among them, the higher the activity level, the more suitable it is to be the target rental node, and the higher the similarity value, the more suitable it is to be the target rental node.

9. The method according to claim 8, characterized in that, The method further includes: Receive the address and license certificate of the target leased node sent by the central scheduling system; Based on the address and license certificate of the target rental node, a computing power rental relationship is established with the target rental node.

10. The method according to claim 9, characterized in that, The license certificate includes the following information: Information on rented resources, the address of the edge node, and the signature of the central scheduling system confirming the legality of this computing power rental.

11. The method according to claim 9, characterized in that, Based on the address and license certificate of the target lease node, a computing power lease relationship is established with the target lease node, including: The local scheduling system of the edge node sends the address of the target leased node, the address of the edge node, and the lease time information to the SDWAN orchestrator and controller, so that the SDWAN orchestrator and controller establish an SDWAN channel between the target leased node and the edge node before the time corresponding to the lease time information. During the time period corresponding to the rental time information, the edge node transmits data and performs rental calculations through the SDWAN channel.

12. The method according to claim 8, characterized in that, The method further includes: A computing power rental request is sent to the scheduling system of the target rental node, so that the target rental node verifies the address of the edge node that sent the computing power rental request according to the address of the edge node recorded in the license certificate, and returns a computing power rental confirmation message after the verification is successful and the resources to be rented are selected.

13. The method according to claim 8, characterized in that, The method further includes: Based on a preset computing power demand algorithm, it is determined whether the edge node needs to rent computing power; Sending a computing power rental request to the central scheduling system includes: When the edge node needs to rent computing power, it sends a computing power rental request to the central scheduling system.

14. The method according to claim 13, characterized in that, Based on a preset computing power demand algorithm, it is determined whether the edge node needs to rent computing power, including: If, within a preset period, the resource utilization rate of the resources in the edge node exceeds a preset utilization rate threshold for a period of time exceeding a preset time threshold, it is determined that the edge node needs to rent computing power.

15. The method according to claim 13, characterized in that, The method further includes: When the edge node needs to rent computing power, an alert is sent to the data owner.

16. The method according to claim 15, characterized in that, The method further includes: Receive the rental fee threshold sent by the data owner.

17. A computing power rental device, characterized in that, The device, applied to a central dispatch system, includes: The rental request receiving module receives computing power rental requests from edge nodes. The computing power rental request includes the address of the edge node, information on the rented computing power resources, rental time information, and rental fee threshold. The cost calculation module is used to calculate the rental cost for each node based on the preset network topology map, the address of the edge node, and the rented computing resources information. The alternative node determination module determines multiple alternative nodes based on the rental fee and the rental fee threshold. The activity calculation module is used to calculate the activity of each candidate node based on the total traffic of each candidate node within a preset period. The similarity calculation module is used to calculate the similarity value between the data used by the candidate nodes and the data required by the edge nodes within a preset time period; The rental node determination module is used to determine the target rental node from the candidate nodes based on the activity level of each candidate node, the rental cost of each candidate node, and the similarity value; wherein, the higher the activity level, the more suitable it is as the target rental node, and the higher the similarity value, the more suitable it is as the target rental node.

18. A computing power rental device, characterized in that, Applied to edge nodes, the device includes: The rental request sending module sends a computing power rental request to the central scheduling system. The computing power rental request includes the address of the edge node, the information of the rented computing power resources, the rental time information, and the rental fee threshold. This allows the central scheduling system to calculate the rental fee for each node based on a preset network topology, the address of the edge node, and the information of the rented computing power resources. Based on the rental fee and the rental fee threshold, the system determines multiple candidate nodes. It also calculates the activity level of each candidate node based on the total traffic of each candidate node within a preset period, calculates the similarity value between the data used by the candidate nodes and the demand data of the edge nodes within a preset period, and determines the target rental node from the candidate nodes based on the activity level of each candidate node, the rental fee of each candidate node, and the similarity value. Among them, the higher the activity level, the more suitable it is to be the target rental node, and the higher the similarity value, the more suitable it is to be the target rental node.

19. A data processing system, characterized in that, The system includes: An edge layer, comprising multiple edge nodes, adopts an integrated storage and computing architecture; the edge nodes are used to send computing power rental requests to the central scheduling system when computing power rental is required, the computing power rental request including the address of the edge node, information on the rented computing power resources, rental time information, and rental fee threshold; The central layer is equipped with a central scheduling system. The central scheduling system is used to receive the computing power rental request, and calculate the rental fee corresponding to each node according to the preset network topology map, the address of the edge node and the rental computing power resource information. Based on the rental fee and the rental fee threshold, multiple candidate nodes are determined. The activity level of each candidate node is calculated according to the total traffic of each candidate node within a preset period. The similarity value between the data used by the candidate nodes and the demand data of the edge nodes within a preset time period is calculated. Based on the activity level of each candidate node, the rental fee of each candidate node and the similarity value, the target rental node is determined from the candidate nodes. Among them, the higher the activity level, the more suitable it is to be the target rental node, and the higher the similarity value, the more suitable it is to be the target rental node.

20. An electronic device, characterized in that, include: processor; as well as Memory for storing the executable instructions of the processor; The processor is configured to execute the computing power rental method according to any one of claims 1-16 by executing the executable instructions.

21. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the computing power rental method according to any one of claims 1-16.

Images