Computing power scheduling method and device and medium

Abstract

The application provides a computing power scheduling method and device and a medium, relates to the technical field of computing power scheduling, and is used for solving the problem of low computing power resource scheduling performance in the prior art. The method comprises the following steps: receiving a computing power scheduling request sent by a computing power demand node; selecting a target computing power resource node for the computing power scheduling request according to a scheduling cost, computing power performance and response capability; and calling the target computing power resource node to complete computing power scheduling for the computing power scheduling request. The scheduling cost, computing power performance and response capability of the computing power resource node are considered simultaneously to perform computing power scheduling, so that the computing power resource scheduling performance is improved, and the demands for various computing power, high response and low cost can be met.

Description

Technical Field

[0001] This invention relates to the field of computing power scheduling technology, and in particular to a computing power scheduling method, apparatus and computer-readable storage medium. Background Technology

[0002] The key to computing power networks is to center computing and base them on the network. Through the deep integration of computing and the network, computing, storage and network resources can be allocated and flexibly scheduled on demand among cloud, network and edge in different business scenarios, making computing power a social-level service like water and electricity, which can be "accessed at one point and used immediately".

[0003] Existing computing power scheduling platforms or solutions use a unified scheduling platform as their technical architecture, with big data centers and supercomputing centers performing unified computing power scheduling. Under the centralized mindset, they lack consideration for the cost and efficiency of computing power scheduling, resulting in poor performance of computing power resource scheduling and failing to meet the demands for diverse, high-response, and low-cost computing power. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to address the above-mentioned shortcomings of the prior art by providing a computing power scheduling method, apparatus and computer-readable storage medium, so as to solve the problem that the prior art lacks consideration of the cost and efficiency of computing power scheduling, resulting in low performance of computing power resource scheduling.

[0005] In a first aspect, the present invention provides a computing power scheduling method, the method comprising:

[0006] Receive computing power scheduling requests sent by nodes that require computing power;

[0007] Select a target computing resource node for the computing power scheduling request based on scheduling cost, computing power performance, and response capability;

[0008] The target computing resource node is invoked to complete the computing power scheduling for the computing power scheduling request.

[0009] Optionally, selecting a target computing resource node for the computing power scheduling request based on scheduling cost, computing power performance, and response capability specifically includes:

[0010] Acquire one or more candidate computing power resource nodes;

[0011] If a candidate computing power resource node is obtained, the candidate computing power resource node shall be used as the target computing power resource node;

[0012] If multiple candidate computing power resource nodes are obtained, the candidate computing power resource node with the highest score is selected as the target computing power resource node according to the following formula:

[0013]

[0014] Among them, F score C represents the score of a candidate computing power resource node that meets the computing power requirement. x C represents the scheduling cost of calling a candidate computing power resource node to complete the computing power scheduling request. v P represents the average scheduling cost of calling the multiple candidate computing power resource nodes to complete the computing power scheduling request. x P represents the computing power performance of a candidate computing power resource node. v R represents the average computing power performance of the multiple candidate computing power resource nodes. x R represents the response capability of a candidate computing resource node. v This represents the average response capability of the multiple candidate computing power resource nodes.

[0015] Optionally, the method is applied to a certain computing power resource node, which is the computing power resource node on the blockchain that is closest to the computing power demand node, and the computing power scheduling request is sent by the computing power demand node after selecting the certain computing power resource node.

[0016] Optionally, acquiring one or more candidate computing power resource nodes specifically includes:

[0017] Obtain the computing power requirement from the computing power scheduling request;

[0018] The system sequentially selects from itself, a preset neighboring region, and the blockchain until it obtains one or more candidate computing power resource nodes whose computing power performance and response capabilities meet the computing power requirements.

[0019] Optionally, before selecting sequentially from itself, a preset neighboring region, and the blockchain until one or more candidate computing power resource nodes whose computing power performance and response capability meet the computing power requirements are obtained, the method further includes:

[0020] Obtain the computing power scheduling smart contract on the blockchain;

[0021] According to the computing power scheduling smart contract, the computing power performance, response capability and geographical location information broadcast by other computing power resource nodes in the preset neighboring area are received and stored in its own first node computing power aggregation table. The preset neighboring area is a region pre-divided on the blockchain with the computing power resource node as the center.

[0022] According to the computing power scheduling smart contract, the computing power performance, response capability and geographical location information of a certain computing power resource node are broadcast to a preset neighboring area, so that other computing power resource nodes in the preset neighboring area can receive the computing power performance, response capability and geographical location information of a certain computing power resource node and store it in their respective second node computing power aggregation tables.

[0023] Optionally, the selection process sequentially involves choosing from the current node, a preset neighboring region, and the blockchain until one or more candidate computing power resource nodes whose computing power performance and response capabilities meet the computing power requirements are obtained. Specifically, this includes:

[0024] Determine whether it is a candidate computing power resource node whose computing power performance and response capability meet the computing power requirements;

[0025] If it is not a candidate computing power resource node, check whether there are one or more candidate computing power resource nodes in the first node computing power aggregation table that meet the computing power requirements in terms of computing power performance and response capability;

[0026] If there are no candidate computing power resource nodes in the first node's computing power aggregation table, then the system will obtain information from other computing power resource nodes outside the preset neighboring area on the blockchain to determine if there are one or more candidate computing power resource nodes whose computing power performance and response capabilities meet the computing power requirements.

[0027] Optionally, if multiple candidate computing power resource nodes are obtained, the method further includes:

[0028] Obtain the geographic location information of the computing power demand node and the plurality of candidate computing power resource nodes;

[0029] The distance between each candidate computing power resource node and the computing power demand node is obtained based on the geographical location information;

[0030] The scheduling cost C of each candidate computing power resource node is obtained based on the distance, historical usage loss, and historical operating costs of each candidate computing power resource node. x .

[0031] Optionally, after the target computing resource node is invoked to complete the computing power scheduling for the computing power scheduling request, the method further includes:

[0032] If the target computing power resource node is itself, then it broadcasts its latest computing power performance and response capability to other computing power resource nodes in the preset neighboring area, so that the other computing power resource nodes in the preset neighboring area update their respective second node computing power aggregation tables.

[0033] If the target computing power resource node is within a preset neighborhood area, the latest computing power performance and response capability broadcast by the target computing power resource node are received to update the first node computing power aggregation table.

[0034] In a second aspect, the present invention provides a computing power scheduling device, comprising:

[0035] The receiving module is used to receive computing power scheduling requests sent by nodes that require computing power.

[0036] The selection module, connected to the receiving module, is used to select a target computing resource node for the computing power scheduling request based on scheduling cost, computing power performance, and response capability.

[0037] The calling module, connected to the selection module, is used to call the target computing power resource node to complete the computing power scheduling for the computing power scheduling request.

[0038] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, it implements the computing power scheduling method described above.

[0039] This invention provides a computing power scheduling method, apparatus, and computer-readable storage medium. By selecting a target computing power resource node for a received computing power scheduling request based on scheduling cost, computing power performance, and response capability, and calling the target computing power resource node to complete the computing power scheduling request, the invention simultaneously considers the scheduling cost, computing power performance, and response capability of the computing power resource node for computing power scheduling, thereby improving the performance of computing power resource scheduling and meeting the demands for diverse computing power needs, high responsiveness, and low cost. Attached Figure Description

[0040] Figure 1 This is a flowchart of a computing power scheduling method according to an embodiment of the present invention;

[0041] Figure 2 This is a schematic diagram of the structure of a computing power scheduling system according to an embodiment of the present invention;

[0042] Figure 3 This is a flowchart of another computing power scheduling method in an embodiment of the present invention;

[0043] Figure 4 This is a flowchart of another computing power scheduling method in an embodiment of the present invention;

[0044] Figure 5 This is a schematic diagram of the structure of a computing power scheduling device in an embodiment of the present invention. Detailed Implementation

[0045] To enable those skilled in the art to better understand the technical solution of the present invention, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0046] It is understood that the specific embodiments and accompanying drawings described herein are merely for explaining the invention and are not intended to limit the invention.

[0047] It is understood that, without conflict, the various embodiments and features in the embodiments of the present invention can be combined with each other.

[0048] It is understood that, for ease of description, only the parts related to the present invention are shown in the accompanying drawings, while the parts unrelated to the present invention are not shown in the drawings.

[0049] It is understood that each unit or module involved in the embodiments of the present invention may correspond to only one entity structure, or may be composed of multiple entity structures, or multiple units or modules may be integrated into one entity structure.

[0050] It is understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of this invention may occur in a different order than that marked in the accompanying drawings.

[0051] It is understood that the flowcharts and block diagrams of this invention illustrate the possible architecture, functions, and operations of systems, apparatuses, devices, and methods according to various embodiments of this invention. Each block in the flowchart or block diagram may represent a unit, module, program segment, or code, containing executable instructions for implementing the specified function. Furthermore, each block or combination of blocks in the block diagram and flowchart can be implemented using a hardware-based system to achieve the specified function, or using a combination of hardware and computer instructions.

[0052] It is understood that the units and modules involved in the embodiments of the present invention can be implemented by software or by hardware. For example, the units and modules can be located in a processor.

[0053] Example 1:

[0054] like Figure 1 As shown, Embodiment 1 of the present invention provides a computing power scheduling method, the method comprising:

[0055] S01, Receive computing power scheduling requests sent by computing power demand nodes;

[0056] S02. Select a target computing resource node for the computing power scheduling request based on scheduling cost, computing power performance, and response capability;

[0057] S03. Call the target computing power resource node to complete the computing power scheduling for the computing power scheduling request.

[0058] Specifically, in this embodiment, the method can be applied to, for example... Figure 2 In the computing power scheduling system shown, such as Figure 2 The computing power scheduling system shown includes multiple distributed computing power resource nodes (peer1-4), each with certain computing power resources (computing power resources AD). These resources include network resources such as 4G / 5G (4 / 5-Generation) mobile networks, IP (Internet Protocol) networks, industrial internet, transmission networks, IoT (Internet of Things) networks, and satellite networks, as well as computing and storage resources such as big data centers, supercomputing centers, and smart terminals. Upon receiving a computing power scheduling request from a computing power demand node, the method selects a target computing power resource node based on scheduling cost, computing power performance, and response capability. The target computing power resource node is then invoked to complete the computing power scheduling. By simultaneously considering the scheduling cost, computing power performance, and response capability of the computing power resource node during scheduling, the system improves computing power resource scheduling performance and can thus meet the demands for diverse, high-response, and low-cost computing power needs.

[0059] Optionally, the method is applied to a certain computing power resource node, which is the computing power resource node on the blockchain that is closest to the computing power demand node, and the computing power scheduling request is sent by the computing power demand node after selecting the certain computing power resource node.

[0060] Specifically, in this embodiment, as Figure 2 In the distributed computing power scheduling system shown, each computing power resource node is also a computing power scheduling node. The computing power scheduling nodes are deployed in the form of blockchain on cloud, network, edge, and terminal operation and management platforms. Each computing power scheduling node connects to a computing power resource and registers its own computing power resource with the blockchain. Each computing power scheduling node can receive computing power scheduling requests. The computing power scheduling requests are sent by the computing power demanding node to the nearest computing power scheduling node to reduce transmission costs.

[0061] Optionally, selecting a target computing resource node for the computing power scheduling request based on scheduling cost, computing power performance, and response capability specifically includes:

[0062] Acquire one or more candidate computing power resource nodes;

[0063] If a candidate computing power resource node is obtained, the candidate computing power resource node shall be used as the target computing power resource node;

[0064] If multiple candidate computing power resource nodes are obtained, the candidate computing power resource node with the highest score is selected as the target computing power resource node according to the following formula:

[0065]

[0066] Among them, F score C represents the score of a candidate computing power resource node that meets the computing power requirement. x C represents the scheduling cost of calling a candidate computing power resource node to complete the computing power scheduling request. v P represents the average scheduling cost of calling the multiple candidate computing power resource nodes to complete the computing power scheduling request. x P represents the computing power performance of a candidate computing power resource node. v R represents the average computing power performance of the multiple candidate computing power resource nodes. x R represents the response capability of a candidate computing resource node. v This represents the average response capability of the multiple candidate computing power resource nodes.

[0067] Optionally, if multiple candidate computing power resource nodes are obtained, the method further includes:

[0068] Obtain the geographic location information of the computing power demand node and the plurality of candidate computing power resource nodes;

[0069] The distance between each candidate computing power resource node and the computing power demand node is obtained based on the geographical location information;

[0070] The scheduling cost C of each candidate computing power resource node is obtained based on the distance, historical usage loss, and historical operating costs of each candidate computing power resource node. x .

[0071] Specifically, in this embodiment, a design formula is used to obtain the computing resource node with the best overall performance in terms of cost and efficiency, with scheduling cost accounting for 60%, computing power performance accounting for 20%, and response capability accounting for 20%; where scheduling cost C x The cost, C, is estimated based on the transmission distance of the computing network, taking into account the depreciation of computing facilities and the operating costs during the computing power usage period. It is calculated by the computing resource nodes. v Then based on multiple C x The average value obtained from the calculation; computing power performance P x R represents the scale and quantity of demands that can be handled, and the responsiveness. x Computing performance (P) represents the ability of computing power to process transactions per unit of time. x and responsiveness R xAs the basic information of computing power resource nodes, it is provided by the computing power resource nodes and maintained on the blockchain.

[0072] Optionally, acquiring one or more candidate computing power resource nodes specifically includes:

[0073] Obtain the computing power requirement from the computing power scheduling request;

[0074] The system sequentially selects from itself, a preset neighboring region, and the blockchain until it obtains one or more candidate computing power resource nodes whose computing power performance and response capabilities meet the computing power requirements.

[0075] Optionally, before selecting sequentially from itself, a preset neighboring region, and the blockchain until one or more candidate computing power resource nodes whose computing power performance and response capability meet the computing power requirements are obtained, the method further includes:

[0076] Obtain the computing power scheduling smart contract on the blockchain;

[0077] According to the computing power scheduling smart contract, the computing power performance, response capability and geographical location information broadcast by other computing power resource nodes in the preset neighboring area are received and stored in its own first node computing power aggregation table. The preset neighboring area is a region pre-divided on the blockchain with the computing power resource node as the center.

[0078] According to the computing power scheduling smart contract, the computing power performance, response capability and geographical location information of a certain computing power resource node are broadcast to a preset neighboring area, so that other computing power resource nodes in the preset neighboring area can receive the computing power performance, response capability and geographical location information of a certain computing power resource node and store it in their respective second node computing power aggregation tables.

[0079] Optionally, the selection process sequentially involves choosing from the current node, a preset neighboring region, and the blockchain until one or more candidate computing power resource nodes whose computing power performance and response capabilities meet the computing power requirements are obtained. Specifically, this includes:

[0080] Determine whether it is a candidate computing power resource node whose computing power performance and response capability meet the computing power requirements;

[0081] If it is not a candidate computing power resource node, check whether there are one or more candidate computing power resource nodes in the first node computing power aggregation table that meet the computing power requirements in terms of computing power performance and response capability;

[0082] If there are no candidate computing power resource nodes in the first node's computing power aggregation table, then the system will obtain information from other computing power resource nodes outside the preset neighboring area on the blockchain to determine if there are one or more candidate computing power resource nodes whose computing power performance and response capabilities meet the computing power requirements.

[0083] Optionally, after the target computing resource node is invoked to complete the computing power scheduling for the computing power scheduling request, the method further includes:

[0084] If the target computing power resource node is itself, then it broadcasts its latest computing power performance and response capability to other computing power resource nodes in the preset neighboring area, so that the other computing power resource nodes in the preset neighboring area update their respective second node computing power aggregation tables.

[0085] If the target computing power resource node is within a preset neighborhood area, the latest computing power performance and response capability broadcast by the target computing power resource node are received to update the first node computing power aggregation table.

[0086] Specifically, in this embodiment, when selecting a target computing power resource node, a computing power information aggregation method is adopted to set up a neighboring area for use as the computing power scheduling node for this time. When the computing power resources of the computing power scheduling node itself cannot meet the computing power scheduling requirements for this time, the computing power resource node that meets the computing power requirements for this time is called first in the neighboring area. The goal is to provide a more efficient, lower carbon, and more economical computing power scheduling scheme for computing power scheduling.

[0087] More specific examples, such as Figure 3 and 4 As shown, the detailed process steps of the method include:

[0088] S1. Set up blockchain nodes; this can be done by, for example... Figure 2 The computing power resource nodes shown are used to build a blockchain, and any computing power resource node is also a blockchain node.

[0089] S2, Synchronous computing power scheduling smart contract; Each blockchain node deploys a computing power calculation contract to synchronously obtain computing power information on the blockchain.

[0090] S3. Upload the computing power information of this node; each computing power resource node broadcasts its own computing power performance, response capability and geographical location information on the blockchain.

[0091] S4. Based on the region, update the computing power information of adjacent nodes to form a computing power aggregation table; each blockchain node receives the computing power performance, response capability and geographical location information of other computing power resource nodes within a certain neighborhood of itself, and forms a computing power aggregation table based on this information to realize regional computing power aggregation and provide a calculation basis for subsequent computing power scheduling.

[0092] S5. Receive computing power scheduling requests sent by computing power demand nodes; the computing power demand nodes send computing power scheduling requests that include their own computing power demand to the nearest node with priority. Sending the request to the nearest computing power resource node can reduce the transmission cost of the request and also facilitate the subsequent selection of computing power resource nodes with low transmission costs (close distance) as the target node for scheduling.

[0093] S6. Complete the computing power scheduling calculation and output the computing power scheduling result; the computing power resource node that receives the computing power scheduling request calculates the computing power scheduling scheme according to the computing power demand. It can perform computing power aggregation and maintenance calculations based on parameters such as computing power type, performance, processing scale, unit price, and network transmission distance, and continuously track the computing power status; specifically including:

[0094] S61. Obtain computing power requirements; Obtain the computing power requirements of the computing power requirement nodes from the computing power scheduling request. This may include parameters such as computing power type, scale, and response efficiency, and may also be divided into computing power performance requirements and response capability requirements.

[0095] S62. Determine whether the current computing power node resources meet the computing power requirements. Since the current node is the node closest to the computing power requirement node in the blockchain, we should give priority to whether the computing power of this node meets the computing power requirements. If so, we should directly take this computing power resource node as the target computing power resource node.

[0096] S63. If not, determine whether the computing power resources of nearby nodes meet the computing power requirements by searching the computing power aggregation table; or if the nodes in the computing power aggregation table still do not meet the computing power requirements, search the computing power aggregation tables of other nodes across multiple nodes to obtain computing power resource nodes that meet the computing power requirements from the entire blockchain.

[0097] S64. If yes, perform computing power scheduling; if a computing power resource node that meets the computing power requirements is obtained, then the node can be scheduled directly; if multiple computing power resource nodes that meet the computing power requirements are obtained, then the optimal computing power resource node is selected by comprehensively considering scheduling cost, computing power performance and response capability. The calculation formula for selecting the optimal node is as described above.

[0098] S7. Perform computing power scheduling, notify the relevant computing power scheduling nodes, and complete the computing power status broadcast; notify the target computing power resource nodes to provide the corresponding computing power resources to the computing power demand nodes, and broadcast the result of this call on the blockchain to avoid conflicts when multiple nodes perform computing power scheduling at the same time.

[0099] S8. Computing power scheduling completed; After this computing power scheduling is completed, the computing power resource status of the called computing power resource node may change, such as changes in storage resources.

[0100] S9. Update the computing power resource status and notify the relevant computing power scheduling nodes to complete the broadcast of the computing power status change; the scheduled computing power resource nodes broadcast their latest computing power status to the blockchain.

[0101] S10. Maintain the computing power aggregation table; the corresponding nodes in the blockchain update their own computing power aggregation table according to the latest computing power status broadcast, so as to maintain the latest computing power information of each computing power resource node.

[0102] In Embodiment 1 of this invention, blockchain technology is used to reliably record computing resources on the chain, achieving distributed computing power scheduling. This provides more efficient computing resource scheduling for computing power demanders. By deploying blockchain at edge nodes in the cloud, network, edge, and terminal, available computing resources can be called upon on a larger scale, idle computing resources can be activated, computing resource efficiency can be improved, and computing power response efficiency can be increased. This fully realizes the decentralized logical architecture of the blockchain, dynamically completing distributed computing power scheduling. All computing resources can be tracked and recorded through the blockchain, achieving reliable sharing of computing resource information on the chain. During computing power invocation, inter-node communication is used to aggregate information on nearby computing power in the region, and a dynamic and efficient computing power scheduling algorithm is designed. The computing power scheduling algorithm calculates the optimal computing power scheduling scheme based on parameters such as computing power scale requirements, computing power timeliness requirements, and computing power cost estimation, completing the on-demand allocation and flexible scheduling of computing power.

[0103] Example 2:

[0104] like Figure 5 As shown, Embodiment 2 of the present invention provides a computing power scheduling device, comprising:

[0105] The receiving module 01 is used to receive computing power scheduling requests sent by computing power demand nodes;

[0106] Selection module 02, connected to the receiving module 01, is used to select a target computing resource node for the computing power scheduling request based on scheduling cost, computing power performance and response capability.

[0107] The calling module 03, connected to the selection module 02, is used to call the target computing power resource node to complete the computing power scheduling for the computing power scheduling request.

[0108] Optionally, the selection module 02 specifically includes:

[0109] Candidate units are used to acquire one or more candidate computing power resource nodes;

[0110] The first target unit is connected to the candidate unit and is used to take the candidate computing power resource node as the target computing power resource node if a candidate computing power resource node is obtained.

[0111] The second target unit, connected to the candidate unit, is used to calculate, if multiple candidate computing power resource nodes are obtained, the candidate computing power resource node with the highest score as the target computing power resource node according to the following formula:

[0112]

[0113] Among them, F score C represents the score of a candidate computing power resource node that meets the computing power requirement. x C represents the scheduling cost of calling a candidate computing power resource node to complete the computing power scheduling request. v P represents the average scheduling cost of calling the multiple candidate computing power resource nodes to complete the computing power scheduling request. x P represents the computing power performance of a candidate computing power resource node. v R represents the average computing power performance of the multiple candidate computing power resource nodes. x R represents the response capability of a candidate computing resource node. v This represents the average response capability of the multiple candidate computing power resource nodes.

[0114] Optionally, the method is applied to a certain computing power resource node, which is the computing power resource node on the blockchain that is closest to the computing power demand node, and the computing power scheduling request is sent by the computing power demand node after selecting the certain computing power resource node.

[0115] Optionally, the candidate unit specifically includes:

[0116] The demand subunit is used to obtain the computing power demand in the computing power scheduling request;

[0117] The selection subunit, connected to the demand subunit, is used to sequentially select from itself, a preset neighboring area, and the blockchain until one or more candidate computing power resource nodes whose computing power performance and response capabilities meet the computing power requirements are obtained.

[0118] Optionally, the device further includes a convergence module, specifically comprising:

[0119] Contract unit, used to obtain computing power scheduling smart contracts on the blockchain;

[0120] The aggregation unit, connected to the contract unit, is used to receive the respective computing power performance, response capability and geographical location information broadcast by other computing power resource nodes in the preset neighboring area according to the computing power scheduling smart contract, and store them in its own first node computing power aggregation table. The preset neighboring area is a region pre-divided on the blockchain with the computing power resource node as the center.

[0121] The broadcast unit, connected to the contract unit, is used to broadcast its own computing power performance, response capability, and geographical location information to a preset neighboring area according to the computing power scheduling smart contract, so that other computing power resource nodes in the preset neighboring area can receive the computing power performance, response capability, and geographical location information of a certain computing power resource node and store it in their respective second node computing power aggregation tables.

[0122] Optionally, the selection subunit specifically includes:

[0123] The first selection subunit is used to determine whether it is a candidate computing power resource node whose computing power performance and response capability meet the computing power requirements.

[0124] The second selection subunit, connected to the first selection subunit, is used to, if it is not a candidate computing power resource node, obtain whether there are one or more candidate computing power resource nodes in the first node computing power aggregation table that meet the computing power requirements in terms of computing power performance and response capability.

[0125] The third selection subunit, connected to the second selection subunit, is used to obtain information from other computing power resource nodes outside the preset neighboring area on the blockchain if there are no candidate computing power resource nodes in the first node computing power aggregation table, and whether there are one or more candidate computing power resource nodes whose computing power performance and response capabilities meet the computing power requirements.

[0126] Optionally, the second target unit further includes:

[0127] The positioning subunit is used to obtain the geographical positioning information of the computing power demand node and the multiple candidate computing power resource nodes;

[0128] A distance subunit, connected to the positioning subunit, is used to obtain the distance between each candidate computing power resource node and the computing power demand node based on the geographical positioning information.

[0129] The cost subunit, connected to the distance subunit, is used to obtain the scheduling cost C of each candidate computing power resource node based on the distance and the historical usage depreciation and historical operating costs of each candidate computing power resource node. x .

[0130] Optionally,

[0131] The broadcasting unit is also used to: if the target computing power resource node is itself, broadcast its latest computing power performance and response capability to other computing power resource nodes in the preset neighboring area, so that the other computing power resource nodes in the preset neighboring area update their respective second node computing power aggregation tables.

[0132] The aggregation unit is further configured to: if the target computing power resource node is within a preset neighborhood area, receive the latest computing power performance and response capability broadcast by the target computing power resource node, so as to update the first node computing power aggregation table.

[0133] The apparatus described in Example 2 corresponds to the method described in Example 1, and Example 2 can be understood in conjunction with Example 1.

[0134] Example 3:

[0135] Embodiment 3 of the present invention provides a computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it implements the computing power scheduling method as described in Embodiment 1.

[0136] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, computer program modules, or other data). Computer-readable storage media include, but are not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory or other memory technologies, CD-ROM (Compact Disc Read-Only Memory), DVD or other optical disc storage, cartridges, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer.

[0137] In addition, the present invention may also provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the computing power scheduling method as described in Embodiment 1.

[0138] The memory is connected to the processor. The memory can be flash memory, read-only memory or other types of memory. The processor can be a central processing unit or a microcontroller.

[0139] Example 3 provides a computer-readable storage medium in which a stored computing program, after being executed, is used to implement computing power scheduling, so as to improve the performance of computing power resource scheduling and thus meet the demands for diverse computing power requirements, high response, and low cost.

[0140] Embodiments 1-3 of the present invention provide a computing power scheduling method, apparatus, and computer-readable storage medium. By selecting a target computing power resource node for a received computing power scheduling request based on scheduling cost, computing power performance, and response capability, and calling the target computing power resource node to complete the computing power scheduling request, the scheduling cost, computing power performance, and response capability of the computing power resource node are considered simultaneously for computing power scheduling, thereby improving the performance of computing power resource scheduling and thus meeting the demands for diverse computing power needs, high response, and low cost.

[0141] It is understood that the above embodiments are merely exemplary implementations used to illustrate the principles of the present invention, and the present invention is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A computing power scheduling method, characterized in that, The method is applied to a specific computing resource node, and the method includes: Receive computing power scheduling requests sent by computing power demand nodes, wherein a certain computing power resource node is the computing power resource node on the blockchain that is closest to the computing power demand node, and the computing power scheduling request is sent by the computing power demand node after selecting the certain computing power resource node; The system receives the computing power performance, response capability and geographical location information broadcast by other computing power resource nodes in a preset neighboring area, and stores it in its own first node computing power aggregation table. The preset neighboring area is a region pre-divided on the blockchain with the computing power resource node as the center. Selecting a target computing resource node for the computing power scheduling request based on scheduling cost, computing power performance, and response capability, specifically including: Obtain the computing power requirement from the computing power scheduling request. It determines whether it is a candidate computing resource node whose computing power performance and response capability meet the computing power requirements. If it is not a candidate computing power resource node, it checks whether there are one or more candidate computing power resource nodes in the first node computing power aggregation table that meet the computing power requirements in terms of computing power performance and response capability. If no candidate computing power resource node is found in the first node's computing power aggregation table, then the computing power aggregation tables of other computing power resource nodes outside the preset neighboring area on the blockchain are searched to determine if one or more candidate computing power resource nodes exist that meet the computing power requirements in terms of computing power performance and response capability. If a candidate computing power resource node is obtained, that candidate computing power resource node shall be used as the target computing power resource node. If multiple candidate computing power resource nodes are obtained, Obtain the geographic location information of the computing power demand node and the plurality of candidate computing power resource nodes; obtain the distance between each candidate computing power resource node and the computing power demand node based on the geographic location information; and obtain the scheduling cost of each candidate computing power resource node based on the distance, historical usage loss, and historical operating costs of each candidate computing power resource node. , The candidate computing power resource node with the highest score is selected as the target computing power resource node according to the following formula: ， in, This represents the score by which a candidate computing power resource node meets the computing power requirement. This represents the scheduling cost of calling a candidate computing power resource node to complete the computing power scheduling request. This represents the average scheduling cost of calling the multiple candidate computing power resource nodes to complete the computing power scheduling request. This represents the computing power performance of a candidate computing power resource node. This represents the average computing power performance of the multiple candidate computing power resource nodes. This indicates the response capability of a candidate computing resource node. This represents the average response capability of the plurality of candidate computing power resource nodes; The target computing power resource node is invoked to complete the computing power scheduling request, and the result of this invocation is broadcast on the blockchain. After the computing power scheduling is completed, the corresponding nodes in the blockchain update their own computing power aggregation table according to the latest computing power status broadcast by the target computing power resource node.

2. The method according to claim 1, characterized in that, Before selecting a target computing resource node for the computing power scheduling request based on scheduling cost, computing power performance, and response capability, the method further includes: Obtain the computing power scheduling smart contract on the blockchain; According to the computing power scheduling smart contract, the computing power performance, response capability and geographical location information of a certain computing power resource node are broadcast to a preset neighboring area, so that other computing power resource nodes in the preset neighboring area can receive the computing power performance, response capability and geographical location information of a certain computing power resource node and store it in their respective second node computing power aggregation tables.

3. The method according to claim 2, characterized in that, After the method calls the target computing resource node to complete the computing power scheduling request, the method further includes: If the target computing power resource node is itself, then it broadcasts its latest computing power performance and response capability to other computing power resource nodes in the preset neighboring area, so that the other computing power resource nodes in the preset neighboring area update their respective second node computing power aggregation tables. If the target computing power resource node is within a preset neighborhood area, the latest computing power performance and response capability broadcast by the target computing power resource node are received to update the first node computing power aggregation table.

4. A computing power scheduling device, characterized in that, The device is a computing resource node, comprising: The receiving module is used to receive computing power scheduling requests sent by computing power demand nodes. The computing power resource node is the computing power resource node on the blockchain that is closest to the computing power demand node. The computing power scheduling request is sent by the computing power demand node after selecting the computing power resource node. The aggregation unit is used to receive the computing power performance, response capability and geographical location information broadcast by other computing power resource nodes in a preset neighboring area, and store them in its own first node computing power aggregation table. The preset neighboring area is a region pre-divided on the blockchain with the computing power resource node as the center. A selection module, connected to the receiving module, is used to select a target computing resource node for the computing power scheduling request based on scheduling cost, computing power performance, and response capability. Specifically, it includes: Obtain the computing power requirement from the computing power scheduling request. It determines whether it is a candidate computing resource node whose computing power performance and response capability meet the computing power requirements. If it is not a candidate computing power resource node, it checks whether there are one or more candidate computing power resource nodes in the first node computing power aggregation table that meet the computing power requirements in terms of computing power performance and response capability. If no candidate computing power resource node is found in the first node's computing power aggregation table, then the computing power aggregation tables of other computing power resource nodes outside the preset neighboring area on the blockchain are searched to determine if one or more candidate computing power resource nodes exist that meet the computing power requirements in terms of computing power performance and response capability. If a candidate computing power resource node is obtained, that candidate computing power resource node shall be used as the target computing power resource node. If multiple candidate computing power resource nodes are obtained, Obtain the geographic location information of the computing power demand node and the plurality of candidate computing power resource nodes; obtain the distance between each candidate computing power resource node and the computing power demand node based on the geographic location information; and obtain the scheduling cost of each candidate computing power resource node based on the distance, historical usage loss, and historical operating costs of each candidate computing power resource node. , The candidate computing power resource node with the highest score is selected as the target computing power resource node according to the following formula: ， in, This represents the score by which a candidate computing power resource node meets the computing power requirement. This represents the scheduling cost of calling a candidate computing power resource node to complete the computing power scheduling request. This represents the average scheduling cost of calling the multiple candidate computing power resource nodes to complete the computing power scheduling request. This represents the computing power performance of a candidate computing power resource node. This represents the average computing power performance of the multiple candidate computing power resource nodes. This indicates the response capability of a candidate computing resource node. This represents the average response capability of the plurality of candidate computing power resource nodes; The calling module, connected to the selection module, is used to call the target computing power resource node to complete the computing power scheduling request and broadcast the result of this call on the blockchain. After the computing power scheduling is completed, the corresponding nodes in the blockchain update their own computing power aggregation table according to the latest computing power status broadcast by the target computing power resource node.

5. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by a processor, implements the computing power scheduling method as described in any one of claims 1-3.

Images