Computing power provider determination method and device, computer equipment and readable storage medium

Abstract

The invention relates to a computing power provider determination method and device, computer equipment and a readable storage medium. The method comprises the following steps: acquiring computing power demand information of a computing power demand side and computing power resource description information of each computing power provider; for each computing power provider, determining a provider reward value of the computing power provider and a demander reward value of the computing power demander according to the computing power demand information and the computing power resource description information of the computing power provider; wherein the provider reward value is a reward value obtained after the computing power provider provides computing power resources meeting the computing power demand information to the computing power demander, and the demander reward value is a reward value obtained after the computing power demander uses the computing power resources provided by the computing power provider; selecting a target provider corresponding to the computing power demander from the computing power providers according to the provider reward value of each computing power provider and each demander reward value of the computing power demander; wherein the target provider is used for providing computing power resources for the computing power demand side.

Description

Technical Field

[0001] The present application relates to the field of computing power technology, and in particular to a method, apparatus, computer device and readable storage medium for determining a computing power provider. Background Art

[0002] With the rapid development of new technologies and chip technologies such as big data, artificial intelligence, and the Internet of Things, computing power is showing a trend of ubiquitous distribution in the cloud, edge, and end. In order to meet the extensive demand for computing power in the digital economy and society, computing power needs to be aggregated with the help of the network to produce universal utility. For computing power to be available on demand like water and electricity, it is necessary to solve the problem of unified management and coordinated scheduling of diverse computing power resources.

[0003] At present, there is no unified standard and specification in the computing power cooperation market, which leads to great subjectivity and uncertainty in the pricing and evaluation of computing power resources. Since the demand side cannot accurately understand the quality and performance of the computing power resources of the supplier, it may lead to the selection of computing power resources that are not suitable for their needs, resulting in resource mismatch, so improvement is urgently needed. Summary of the invention

[0004] Based on this, it is necessary to provide a method, device, computer equipment and readable storage medium for determining a computing power provider that can optimize the accuracy of computing power matching in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for determining a computing power provider, the method comprising:

[0006] Obtain computing power demand information of computing power demanders and computing power resource description information of each computing power provider;

[0007] For each computing power provider, determine the provider reward value of the computing power provider and the demander reward value of the computing power demander based on the computing power demand information and the computing power resource description information of the computing power provider; the provider reward value is the reward value obtained by the computing power provider after providing the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider;

[0008] According to the provider reward value of each computing power provider and the demander reward value of each computing power demander, a target provider corresponding to the computing power demander is selected from each computing power provider; wherein the target provider is used to provide computing power resources to the computing power demander.

[0009] In one embodiment, the computing power resource description information includes resource information of available computing power resources under different computing power types, and the resource information of available computing power resources under each computing power type includes the resource amount, deployment location information and unit computing power resource value of the available computing power resources under the computing power type;

[0010] According to the computing power demand information and the computing power resource description information of the computing power provider, the provider reward value of the computing power provider is determined, including:

[0011] From different computing power types, select the target computing power type that matches the computing power type requirement in the computing power requirement information;

[0012] When it is determined that the amount of available computing resources under the target computing power type meets the resource amount requirement in the computing power requirement information, select the target computing power resources that meet the resource amount requirement from the available computing power resources under the target computing power type;

[0013] The provider reward value of the computing power provider is determined based on the resource quantity, deployment location information and unit computing power resource value of the target computing power resources.

[0014] In one of the embodiments, the resource information also includes a computing power performance indicator of each unit computing power resource in the available computing power resources under the computing power type;

[0015] Select the target computing resources that meet the resource requirements from the available computing resources under the target computing type, including:

[0016] Select a candidate computing resource whose computing power performance index meets the computing power performance requirement in the computing power requirement information from each unit computing power resource included in the available computing power resources under the target computing power type;

[0017] From the candidate computing resources, select the target computing resources that meet the resource requirements.

[0018] In one embodiment, the resource information further includes a computing power cost indicator of each unit computing power resource in the available computing power resources under the computing power type; the computing power cost indicator of each unit computing power resource includes the total amount of data that can be processed by the unit computing power resource, the computing cost of processing the total amount of data, and the communication cost;

[0019] The provider reward value of the computing power provider is determined based on the resource volume, deployment location information and unit computing power resource value of the target computing power resources, including:

[0020] Determine the computing power processing capacity of the target computing power resources based on the ratio between the resource amount of the target computing power resources and the resource amount of the available computing power resources of the computing power provider;

[0021] The provider reward value of the computing power provider is determined based on the computing power cost indicators, deployment location information and unit computing power resource value of each unit computing power resource in the target computing power resources, as well as the computing power processing capacity of the target computing power resources.

[0022] In one embodiment, determining the demander reward value of the computing power demander according to the computing power demand information and the computing power resource description information of the computing power provider includes:

[0023] The reward value for the computing power demander is determined based on the computing power processing capacity, resource volume and unit computing power resource value of the target computing power resources.

[0024] In one embodiment, according to the provider reward value of each computing power provider and each demander reward value of the computing power demander, selecting a target provider corresponding to the computing power demander from each computing power provider includes:

[0025] Determine the set of providers based on the computing power providers corresponding to the reward values ​​of each computing power demander;

[0026] The computing power provider with the largest demander reward value in the provider set is selected as the candidate provider;

[0027] determining whether the provider reward value of the candidate provider is greater than or equal to the expected reward value;

[0028] If yes, the candidate provider will be used as the target provider corresponding to the computing power demander;

[0029] If not, the computing power provider corresponding to the largest demander reward value is selected from the providers other than the candidate provider in the provider set as the updated candidate provider, and based on the updated candidate provider, the step of determining whether the provider reward value of the candidate provider is greater than or equal to the expected reward value is returned to execute.

[0030] In a second aspect, the present application further provides a computing power provider determination device, the device comprising:

[0031] The acquisition module is used to obtain the computing power demand information of the computing power demander and the computing power resource description information of each computing power provider;

[0032] The reward calculation module is used to determine the provider reward value of the computing power provider and the demander reward value of the computing power demander for each computing power provider according to the computing power demand information and the computing power resource description information of the computing power provider; wherein the provider reward value is the reward value obtained by the computing power provider after providing the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider;

[0033] The computing power provider determination module is used to select the target provider corresponding to the computing power demander from among the computing power providers based on the provider reward value obtained by each computing power provider when meeting the computing power demand information, and the demander reward value obtained by the computing power demander when using the computing power resources provided by each computing power provider; the target provider is used to provide computing power services to the computing power demander.

[0034] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0035] Obtain computing power demand information of computing power demanders and computing power resource description information of each computing power provider;

[0036] For each computing power provider, determine the provider reward value of the computing power provider and the demander reward value of the computing power demander based on the computing power demand information and the computing power resource description information of the computing power provider; the provider reward value is the reward value obtained by the computing power provider after providing the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider;

[0037] According to the provider reward value of each computing power provider and the demander reward value of each computing power demander, a target provider corresponding to the computing power demander is selected from each computing power provider; wherein the target provider is used to provide computing power resources to the computing power demander.

[0038] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0039] Obtain computing power demand information of computing power demanders and computing power resource description information of each computing power provider;

[0040] For each computing power provider, determine the provider reward value of the computing power provider and the demander reward value of the computing power demander based on the computing power demand information and the computing power resource description information of the computing power provider; the provider reward value is the reward value obtained by the computing power provider after providing the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider;

[0041] According to the provider reward value of each computing power provider and the demander reward value of each computing power demander, a target provider corresponding to the computing power demander is selected from each computing power provider; wherein the target provider is used to provide computing power resources to the computing power demander.

[0042] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0043] Obtain computing power demand information of computing power demanders and computing power resource description information of each computing power provider;

[0044] For each computing power provider, determine the provider reward value of the computing power provider and the demander reward value of the computing power demander based on the computing power demand information and the computing power resource description information of the computing power provider; the provider reward value is the reward value obtained by the computing power provider after providing the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider;

[0045] According to the provider reward value of each computing power provider and the demander reward value of each computing power demander, a target provider corresponding to the computing power demander is selected from each computing power provider; wherein the target provider is used to provide computing power resources to the computing power demander.

[0046] The above-mentioned computing power provider determination method, device, computer equipment and readable storage medium, this application can establish a comprehensive information database by obtaining the specific demand information of the computing power demander and the resource description information of each computing power provider. Utilizing this information, the computing power resources of the demander and the provider can be accurately matched, resource mismatch can be reduced, matching efficiency can be improved, and provider reward values ​​and demander reward values ​​can be set to provide clear incentives for both parties of the computing power cooperation. The provider reward value encourages the computing power provider to provide high-quality and high-performance computing power resources to meet the diverse needs of the demander, while the demander reward value encourages the demander to use computing power resources reasonably and improve resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 A diagram of an application environment of a method for determining a computing power provider in an embodiment;

[0049] Figure 2 A flowchart of a method for determining a computing power provider in one embodiment;

[0050] Figure 3 A flowchart of a step of determining a provider reward value of a computing power provider in one embodiment;

[0051] Figure 4 A flowchart of a step of determining a provider reward value of a computing power provider in another embodiment;

[0052] Figure 5 A flowchart of a method for determining a computing power provider in another embodiment;

[0053] Figure 6 It is a structural block diagram of a computing power provider determination device in one embodiment;

[0054] Figure 7 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0056] In an exemplary embodiment, Figure 1 As shown, this embodiment provides an application environment of a method for determining a computing power provider, and the application environment includes a computing power cooperation platform 10, at least one computing power demander 20 and multiple computing power providers 30.

[0057] In an exemplary embodiment, Figure 2 As shown, a method for determining a computing power provider is provided, and the method is applied to Figure 1 Taking the computing power cooperation platform 10 in the example as an example, the method includes:

[0058] S201, obtaining computing power demand information of the computing power demander and computing power resource description information of each computing power provider.

[0059] Among them, computing power demand information refers to the specific information about the required computing power resources reported or proposed by the computing power demander when using computing power cooperation. This information usually includes but is not limited to the following aspects: (1) Computing power resource type requirements: the type of computing power resources that the computing power demander hopes to obtain, such as central processing unit (CPU) computing power, graphics processing unit (GPU) computing power, field programmable gate array (FPGA) computing power, etc. These different types of computing power resources have differences in computing power, applicable scenarios, etc. (2) Computing power resource scale requirements: the computing power resource scale required by the computing power demander, including the required computing power size (such as computing power, storage capacity, etc.) and duration (such as the time period during which the task needs to run). (3) Performance index requirements: the computing power demander's performance index requirements for computing power resources, such as processing speed, response time, accuracy, etc. These requirements are usually related to specific computing tasks or application scenarios. (4) Bidding information: the price or cost that the computing power demander is willing to pay to obtain the required computing power resources.

[0060] Among them, the computing power resource description information of the computing power provider refers to the specific information about the computing power resources it owns that is reported or provided by the computing power provider when registering or joining the computing power cooperation platform.

[0061] It is understandable that in order to meet the computing power demand information of the computing power demander, the computing power resource description information should correspond to the computing power demand information. Optionally, the computing power resource description information includes: (1) Computing power resource ID and type: the unique identifier (ID) of the computing power resource, which is used to distinguish and track different computing power resources in computing power cooperation. (2) The type of computing power resource, such as CPU computing power, GPU computing power, FPGA computing power, artificial intelligence (AI) chip computing power, etc., as well as specific models or specifications. (3) Real-time available resource status: the current real-time available status of the computing power resource, including the size of idle computing power, the size of occupied computing power, the estimated available time, etc. (4) Resource performance and specifications: specific performance indicators of computing power resources, such as computing power (FLOPS, TOPS, etc.), storage capacity, network bandwidth, etc.; resource specification information, such as the number of CPU cores, main frequency, memory size, disk capacity, etc., as well as the GPU video memory size, number of stream processors, etc. (5) Resource quotation and billing method: The price or cost set by the computing power provider for the computing power resources, the billing method and pricing model provided, such as hourly billing, task-based billing, resource quantity-based billing, etc., as well as possible discounts, offers or package options.

[0062] In addition, the computing power resource description information may also include additional information and services that the computing power provider may provide, such as technical support, resource monitoring, security policies, data protection, etc., as well as the computing power provider's reputation rating, historical cooperation records and user evaluations, to help computing power demanders evaluate the provider's reliability and service quality.

[0063] S202, for each computing power provider, determine the provider reward value of the computing power provider and the demander reward value of the computing power demander based on the computing power demand information and the computing power resource description information of the computing power provider.

[0064] Among them, the provider reward value is the reward value obtained by the computing power provider after providing computing power resources that meet the computing power demand information to the computing power demander.

[0065] It is understandable that the provider reward value represents the reward or return obtained by the computing power provider after meeting the needs of the computing power demander. The provider reward value mainly comes from the incentive mechanism within the computing power cooperation platform. When the computing power demander proposes computing power demand information, such as computing power resource type, scale, performance indicators, etc., the computing power cooperation will match this information with the computing power resource description information of the computing power provider. If the resources of the computing power provider can meet the requirements of the demander, the two parties can reach a cooperation, and the computing power provider can obtain the corresponding reward value after providing the corresponding computing power resources.

[0066] Exemplarily, the calculation basis of the provider's reward value mainly includes the following aspects: (1) Resource matching degree: the matching degree between the resources of the computing power provider and the needs of the computing power demander. The higher the matching degree, the more the computing power provider's resources can meet the requirements of the demander, and therefore the reward value obtained is correspondingly higher. (2) Resource quality: the quality of the computing power resources provided by the computing power provider, including computing performance, stability, reliability, etc. The higher the resource quality, the greater the contribution of the computing power provider in the cooperation, and therefore the more reward value obtained. (3) Cooperation price: the price set by the computing power provider in the cooperation. In the market competition, a reasonable price strategy can attract more computing power demanders, thereby increasing the amount of cooperation and improving the benefits of the computing power provider. (4) Market competition situation: The competition situation in the computing power cooperation market will also affect the calculation of the provider's reward value. In the fierce market competition, the computing power provider may need to lower prices, improve resource quality or optimize services to attract demanders, which may result in a relatively low reward value. However, in a market with weak competition, the computing power provider may obtain a higher reward value.

[0067] Among them, the demander's reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider.

[0068] It is understandable that the demander's reward value represents the reward or benefit obtained by the computing power demander after using the computing power resources provided by the computing power provider. The demander's reward value mainly comes from the settlement mechanism within the computing power cooperation platform. When the computing power demander obtains the required computing power resources through cooperation and successfully completes the computing task, the corresponding reward value will be calculated for the computing power demander based on factors such as the actual use of computing power resources, the quality of task completion, and the service performance of the computing power provider.

[0069] For example, this reward value may include multiple aspects, such as the efficiency of computing resource utilization, the speed of completing computing tasks, the accuracy of data processing, etc. These factors will affect the actual benefits of the computing power demander in the cooperation, and therefore will also become an important basis for the calculation of the reward value.

[0070] For example, the calculation basis of the demander's reward value mainly includes the following aspects: (1) Computing power resource usage: the computing power demander's efficiency of using computing power resources, the length of time of use, etc. These factors reflect the computing power demander's actual demand for resources and utilization, and are one of the important bases for calculating the reward value. (2) Computing task completion: the quality and speed of the computing tasks completed by the computing power demander through computing power resources. These factors directly reflect the effect and value of computing power resources in solving practical problems, and are also important indicators for evaluating the benefits of computing power demanders. (3) The service performance of the computing power provider: the computing power provider's service attitude, response speed and resource stability in the cooperation.

[0071] S203, selecting a target provider corresponding to the computing power demander from among the computing power providers according to the provider reward value of each computing power provider and the demander reward value of each computing power demander.

[0072] Among them, the target provider is used to provide computing power resources to the computing power demander.

[0073] Optionally, after obtaining the provider reward values ​​of each computing power provider and the demander reward values ​​of the computing power demander, a comprehensive analysis and comparison will be conducted. This process may involve considerations in multiple dimensions, such as the matching degree of computing power resources, price advantages, service quality, historical cooperation records, etc. By comprehensively considering these factors, a computing power provider that best suits the computing power demander will be selected as the target provider.

[0074] In one feasible method, an evaluation model may be constructed, with the provider reward value of each computing power provider and the demander reward value of each computing power demander as inputs for constructing the model. The output of the constructed model is to select the target provider corresponding to the computing power demander from among the computing power providers.

[0075] In another possible implementation, an optimization algorithm can be selected that can maximize the demander's benefit when selecting the target provider corresponding to the computing power demander from various computing power providers, and ensure that the provider reward of the computing power provider that provides computing power resources also meets the corresponding reward threshold, so that both the computing power demander and the computing power provider are willing to cooperate.

[0076] Once the target provider is determined, the relevant information will be notified to the computing power demander and computing power provider. The computing power demander can further communicate and negotiate with the target provider to reach a cooperation and obtain the required computing power resources.

[0077] The above-mentioned method for determining computing power providers, this application can establish a comprehensive information database by obtaining the specific demand information of the computing power demander and the resource description information of each computing power provider. Utilizing this information, the computing power resources of the demander and the provider can be accurately matched, resource mismatch can be reduced, matching efficiency can be improved, and provider reward values ​​and demander reward values ​​can be set to provide clear incentives for both parties of the computing power cooperation. The provider reward value encourages the computing power provider to provide high-quality and high-performance computing power resources to meet the diverse needs of the demander, while the demander reward value encourages the demander to use computing power resources reasonably and improve resource utilization.

[0078] In an exemplary embodiment, the computing power resource description information includes resource information of available computing power resources under different computing power types, and the resource information of available computing power resources under each computing power type includes the resource quantity, deployment location information and unit computing power resource value of the available computing power resources under the computing power type.

[0079] Among them, the available computing power resources include multiple unit computing power resources, and each unit computing power resource corresponds to corresponding deployment location information and unit computing power resource value.

[0080] It is understandable that the deployment location information has an important impact on the response speed of computing power requirements, network latency, data transmission costs, etc. The cost or value of each unit of computing power resources can be an economic cost (such as the price per core per hour), a performance indicator (such as the price per FLOPS), or other metrics. The available computing power resources are subdivided into multiple unit computing power resources, each of which has independent deployment location information and unit computing power resource value. This subdivision makes the management and scheduling of computing power resources more flexible, and computing power resources can be dynamically allocated and adjusted according to actual needs.

[0081] like Figure 3 As shown, according to the computing power demand information and the computing power resource description information of the computing power provider, the provider reward value of the computing power provider is determined, including:

[0082] S301: Select a target computing power type that matches the computing power type requirement in the computing power requirement information from different computing power types.

[0083] Optionally, the computing power type requirement is extracted from the computing power requirement information, which is usually one or more specific computing power types, such as CPU computing power, GPU computing power, FPGA computing power, etc. Then, the computing power resource description information provided by the computing power provider is traversed to find all computing power types that match the computing power type requirements in the computing power requirement information. These matching computing power types are candidate target computing power types.

[0084] S302: When it is determined that the amount of available computing resources under the target computing power type meets the resource amount requirement in the computing power requirement information, select a target computing power resource that meets the resource amount requirement from the available computing power resources under the target computing power type.

[0085] Optionally, for each candidate target computing power type, check whether the amount of available computing power resources under it meets the resource amount requirements in the computing power requirement information. This usually involves comparing two values: the resource amount requirements in the computing power requirement information and the total amount of available computing power resources or the remaining amount of available computing power resources under the target computing power type. If the amount of available computing power resources under a certain target computing power type meets the requirements, a sufficient number of computing power resources are selected from this type to meet the resource amount requirements in the computing power requirement information. These selected computing power resources are the target computing power resources.

[0086] S303, determining the provider reward value of the computing power provider according to the resource quantity, deployment location information and unit computing power resource value of the target computing power resources.

[0087] In one possible implementation, the basic resource value is calculated based on the resource quantity of the target computing power resources (such as the number of CPU cores, the number of GPUs, the memory size, etc.) and the unit computing power resource value (such as the price per core per hour, the price per GB memory per hour, etc.). Basic resource value = resource quantity × unit computing power resource value.

[0088] Next, evaluate the impact of the deployment location information on the actual use and cost of computing resources. If the computing power demander and computing power provider are located in the same region or the network latency is low, set a higher adjustment factor (such as 1.1 or higher), indicating that the resource value increases. If the computing power demander and computing power provider are geographically far away or the network latency is high, set a lower adjustment factor (such as 0.9 or lower), indicating that the resource value decreases. The specific value of the adjustment factor can be determined based on actual conditions and negotiation between the two parties. Use the deployment location adjustment factor to adjust the basic resource value. Adjusted resource value = basic resource value × deployment location adjustment factor.

[0089] Finally, the provider reward value of the computing power provider is determined based on the adjusted resource value. This reward value can be the adjusted resource value itself, or it can be a value determined after further considering other factors (such as market supply and demand, the cost of the computing power provider, etc.).

[0090] In another possible implementation, the resource information also includes computing power performance indicators of each unit of computing power resources in the available computing power resources under the computing power type.

[0091] It is understandable that computing power performance indicators refer to specific indicators used to measure and evaluate the computing power of each unit of computing power resources. These indicators generally reflect the efficiency and ability of computing power resources in performing various computing tasks.

[0092] The following are some common computing performance indicators: (1) Floating Point Operations Per Second (FLOPS): It stands for the number of floating point operations per second and is an important indicator for measuring the performance of the processor in floating point operations. (2) Integer operation performance: Integer operation performance measures the number of integer operations that the processor can perform per second. (3) Memory bandwidth: Memory bandwidth refers to the data transmission speed between the computer memory and the processor. (4) Storage performance: Storage performance includes the read and write speed of the hard disk and the performance of the solid state drive (SSD). (5) Parallel processing capability: Parallel processing capability refers to the ability to process multiple computing tasks simultaneously. (6) Computing power accuracy: Computing power accuracy refers to the numerical accuracy used in the calculation process. (7) It is divided into floating point calculation (half precision, single precision, double precision) and integer calculation. (8) Hash rate: Hash rate refers to the number of hash function calculations that a computer system can perform per second. (9) BOPs (Billions of Operations Per Second): BOPs refers to the number of billion operations that can be performed per second. (10) Energy efficiency ratio: Energy efficiency ratio refers to the ratio between the energy consumed by a computer when performing computing tasks and the computing performance provided.

[0093] Specifically, one or more key performance indicators may be selected from the above-mentioned common computing power performance indicators.

[0094] Accordingly, a target computing power resource that meets the resource quantity requirement is selected from the available computing power resources under the target computing power type, including: selecting a candidate computing power resource whose computing power performance indicator meets the computing power performance requirement in the computing power requirement information from each unit computing power resource included in the available computing power resources under the target computing power type; and selecting a target computing power resource that meets the resource quantity requirement from the candidate computing power resources.

[0095] It is understandable that for each candidate target computing power type, first check whether the amount of available computing power resources under it meets the resource amount requirements in the computing power demand information. At the same time, check whether the computing power performance indicators of these resources meet the computing power performance requirements in the computing power demand information. The computing power performance indicators may include computing speed, memory size, bandwidth, latency, etc., depending on the application scenarios and requirements of the computing power demander.

[0096] In one achievable manner, from all available computing power resources under the target computing power type, those computing power resources whose computing power performance indicators meet the computing power demand information are screened out as candidate computing power resources. If it is not possible to screen out those computing power resources whose computing power performance indicators meet the computing power demand information, a request can be made to the computing power demander to change the computing power type. If the computing power demander agrees to the request, those computing power resources whose computing power performance indicators meet the computing power demand information can be screened out from other computing power types as candidate computing power resources.

[0097] Furthermore, among the candidate computing resources, multiple unit computing resources may need to be combined to meet the resource quantity requirements in the computing demand information. This can be achieved by accumulating the resource quantities of the candidate computing resources while ensuring that the combined resources still meet the requirements in terms of computing performance.

[0098] Optionally, if there are multiple combinations, you can choose based on other factors (such as cost, deployment location, resource availability, etc.). Select an optimal combination from all possible combinations as the target computing resource. This optimal combination should not only meet the resource quantity requirements, but also meet or exceed the computing power demander's requirements in terms of computing power performance.

[0099] In an exemplary embodiment, the resource information also includes a computing power cost indicator of each unit computing power resource in the available computing power resources under the computing power type.

[0100] Among them, the computing cost indicators of each unit of computing power resources include the total amount of data that the unit computing power resources can process, the computing cost of processing the total amount of data, and the communication cost.

[0101] It is understandable that the total amount of data that can be processed by a unit of computing power resources refers to the amount of data that can be processed or completed by each unit of computing power resources within a specific time (such as one second, one minute, etc.). It reflects the processing capacity or throughput of computing power resources.

[0102] The computational cost of processing the total amount of data refers to the computing resources or energy consumed by computing resources to process the above total amount of data. This may include CPU / GPU usage time, memory usage, energy consumption, etc. Computational cost is one of the important indicators to measure the efficiency of computing resources.

[0103] Communication cost of processing total data volume: refers to the cost or delay incurred by network communication for processing data. This includes bandwidth consumption for data transmission, network delay, packet loss rate, etc. For applications that require frequent data transmission or are sensitive to network delay, communication cost is an important consideration.

[0104] On this basis, if Figure 4As shown, according to the resource quantity, deployment location information and unit computing resource value of the target computing resource, the provider reward value of the computing resource provider is determined, including:

[0105] S401, determining the computing power processing capacity of the target computing power resources according to the ratio between the resource amount of the target computing power resources and the resource amount of the available computing power resources of the computing power provider.

[0106] Among them, the ratio between the amount of target computing resources and the amount of available computing resources of the computing power provider can indirectly indicate how many idle resources there are in the available computing power resources of the computing power provider. It is understandable that the amount of idle resources will indirectly affect the processing capacity of the unit computing power. When the amount of idle resources is large, it means that the computing power provider has more resources that can be flexibly allocated to computing power consumers, so the processing capacity of the unit computing power may be higher; conversely, when the amount of idle resources is small, the computing power provider may not be able to meet the needs of all computing power consumers in a timely manner, resulting in a decrease in the processing capacity of the unit computing power.

[0107] Therefore, the greater the ratio between the target computing resources and the available computing resources of the computing power provider, the greater the computing power processing capacity of the target computing resources; conversely, the smaller the ratio between the target computing resources and the available computing resources of the computing power provider, the smaller the computing power processing capacity of the target computing resources.

[0108] S402, determining the provider reward value of the computing power provider according to the computing power cost index, deployment location information and unit computing power resource value of each unit computing power resource in the target computing power resources, and the computing power processing capacity of the target computing power resources.

[0109] Optionally, for each unit of computing power, its efficiency or utility E can be calculated i , E i It can be the ratio of the amount of data processed to the computational cost and communication cost, or other appropriate measurement methods. According to the deployment location information, adjust the efficiency or utility of each unit of computing power resources. For example, if the deployment location of a unit of computing power resources is particularly beneficial to the computing power demander, its efficiency value can be increased: E i ′=E i ×L i ; Multiply the adjusted efficiency or utility by the value of the unit computing power resource to obtain the initial reward value R of each unit computing power resource i =E i ′×V i ; The initial reward value is further adjusted based on the computing power processing capacity P of the target computing power resource. For example, if the computing power processing capacity is high, the reward value can be increased; otherwise, it can be reduced. This can be achieved through a weighting factor related to P: R i ′=Ri ×f(P); add up the adjusted reward values ​​of all unit computing resources to get the provider reward value R of the computing power provider: R= i=1 ∑nR i ′.

[0110] In an exemplary embodiment, the demander reward value of the computing power demander is determined based on the computing power demand information and the computing power resource description information of the computing power provider, including: the computing power provider determines the demander reward value of the computing power demander based on the computing power processing capacity of the target computing power resources, the resource volume of the target computing power resources and the computing power value of the unit computing power resources in the target computing power resources.

[0111] Optional, the total computing power value T of the target computing power resource v It can be obtained by multiplying the computing power value V of the unit computing power resource by the resource quantity Q: v =Q×V, where V can be determined based on market supply and demand, the cost of computing power providers, the willingness to pay of computing power demanders, etc. The computing power processing capacity P reflects the efficiency or performance of the target computing power resources.

[0112] In order to include the computing power into the calculation of the reward value, a weighting factor f(P) related to P can be introduced. f(P) can be a linear function, exponential function, or other form of function of P, depending on the application scenario and actual needs. For example, if you want the reward value to be proportional to the computing power, you can choose f(P) = kP (where k is a constant).

[0113] Finally, multiply the total computing power value Tv by the weighting factor f(P) to obtain the demand-side reward value Rd of the computing power demander: Rd=Tv×f(P).

[0114] In an exemplary embodiment, Figure 5 As shown, according to the provider reward value of each computing power provider and the demander reward value of each computing power demander, a target provider corresponding to the computing power demander is selected from each computing power provider, including:

[0115] S501, determining a provider set according to the computing power providers corresponding to the reward values ​​of each computing power demander.

[0116] Optionally, the purpose of determining the provider set based on the computing power providers corresponding to the reward values ​​of each computing power demander is to screen out the group of computing power providers that can meet the requirements of the computing power demanders and are of the highest value to them in terms of the reward dimension.

[0117] S502: The computing power provider with the largest reward value from the demander in the provider set is selected as a candidate provider.

[0118] Optionally, the demander's reward values ​​of each computing power provider in the provider set are compared to obtain the computing power provider with the largest reward value as the candidate provider. The largest reward value indicates that cooperation with this provider can obtain the required computing power resources at the most reasonable cost, thereby maximizing cost-effectiveness. The computing power resources provided by this provider are highly matched with the computing power demander's needs in terms of performance, specifications, availability, etc., and can meet or exceed the expectations of the demander. The computing power provider has good performance in service quality, such as fast response speed, strong fault recovery capability, professional technical support, etc., which can improve the satisfaction and trust of the demander.

[0119] S503, determining whether the provider reward value of the candidate provider is greater than or equal to the expected reward value; if so, executing S504, if not, executing S505.

[0120] It is understandable that the expected reward value refers to the minimum reward level that the computing power provider is willing to participate in the computing power cooperation. The purpose of determining whether the provider reward value of the candidate provider is greater than or equal to the expected reward value is to determine whether the candidate computing power provider can guarantee the minimum reward when participating in the computing power cooperation. When the computing power cooperation platform serves the computing power demanders and computing power providers, it is necessary to ensure that the rewards of both parties can meet expectations.

[0121] S504: Use the candidate provider as the target provider corresponding to the computing power demander.

[0122] It can be understood that if the provider corresponding to the candidate provider establishes a reward greater than or equal to the expected reward, the candidate provider will be used as the target provider corresponding to the computing power demander.

[0123] S505: If not, then from the providers other than the candidate provider in the provider set, the computing power provider corresponding to the largest demander reward value is used as the updated candidate provider, and based on the updated candidate provider, return to execute S503.

[0124] For example, the provider reward value of candidate provider A (i.e., the fee that A requires X to pay) is 90 units, and the expected reward value (i.e., the maximum fee that X is willing to pay) is 100 units.

[0125] The judgment result of S503 is: the provider reward value of A (90 units) is greater than the expected reward value (100 units). Therefore, step S505 is executed.

[0126] S505 execution: exclude candidate provider A from the provider set {A, B, C, D}, and the remaining ones are {B, C, D}.

[0127] Among the remaining providers, the demand-side reward values ​​of each are compared: the demand-side reward value of provider B is 90 units. The demand-side reward value of provider C is 110 units. The demand-side reward value of provider D is 80 units.

[0128] Select provider C (110 units) with the largest demander reward value as the updated candidate provider. Re-execute S503: The provider reward value of candidate provider C (i.e., the fee that C requires X to pay) is 80 units. The expected reward value is still 100 units. Judgment result: C's provider reward value (110 units) is greater than the expected reward value (100 units). Therefore, execute step S504 and select D as the target provider corresponding to computing power demander X.

[0129] In an exemplary embodiment, in the above S203, an evaluation model may be constructed, and the provider reward value of each computing power provider and the demander reward value of each computing power demander are used as inputs for constructing the model. The output of the constructed model is to select the target provider corresponding to the computing power demander from each computing power provider. The specific process is as follows:

[0130] In this embodiment, the evaluation model includes a fuzzy comprehensive evaluation sub-model, a hierarchical analysis method sub-model, a grey relational analysis sub-model and a fusion evaluation model; wherein:

[0131] First, obtain the reward values ​​of each demander of computing power (that is, the fees or rewards they are willing to pay for obtaining computing power resources).

[0132] Secondly, (1) the execution process of the fuzzy comprehensive evaluation sub-model is as follows: Determine the evaluation factors: According to the actual situation of computing power cooperation, determine the key factors that affect the selection of the target provider, such as computing power performance, stability, response speed, service quality, etc. Construct the fuzzy evaluation matrix: Perform fuzzy evaluation on each computing power provider according to the above evaluation factors to obtain the fuzzy evaluation matrix. Calculate the fuzzy comprehensive evaluation results: Use appropriate fuzzy operation methods to calculate the fuzzy evaluation matrix and obtain the fuzzy comprehensive evaluation results of each computing power provider.

[0133] Then, the execution process of (2) the AHP sub-model is as follows: Establish a hierarchical structure: decompose the problem of selecting computing power providers into multiple levels, such as the target level, the criterion level, the solution level, etc. Construct a judgment matrix: For each level, construct a judgment matrix to represent the relative importance of each evaluation factor. Calculate the weight vector: By solving the eigenvector of the judgment matrix or using other methods, calculate the weight vector of each evaluation factor.

[0134] Then, the execution process of (3) the grey correlation analysis sub-model is as follows: Determine the reference sequence: Take the expectations or standards of the computing power demander as the reference sequence. Construct the comparison sequence: Take the relevant indicators of each computing power provider (such as the provider reward value, computing power performance, etc.) as the comparison sequence. Calculate the correlation degree: Use the grey correlation analysis method to calculate the correlation degree between each comparison sequence and the reference sequence.

[0135] Finally, (4) the execution process of the fusion evaluation model is as follows: Fusion sub-model results: Fusion the results of the fuzzy comprehensive evaluation sub-model, the hierarchical analysis method sub-model, and the grey correlation analysis sub-model. The fusion method can be weighted average, fuzzy integration, evidence theory, etc. Constructing a fusion evaluation matrix: Based on the fusion results, a fusion evaluation matrix is ​​constructed, in which each element represents the comprehensive evaluation value of the corresponding computing power provider. Sorting and selection: Sort the elements in the fusion evaluation matrix, and select the computing power provider with the highest comprehensive evaluation value as the target provider. Output results: The selected target provider and its related information (such as provider reward value, computing power performance, etc.) are used as the output of the model.

[0136] It should be understood that, although the steps in the flowcharts involved in the above embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0137] Based on the same inventive concept, the embodiment of the present application also provides a computing power provider determination device for implementing the computing power provider determination method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more computing power provider determination device embodiments provided below can refer to the limitations of the computing power provider determination method above, and will not be repeated here.

[0138] In an exemplary embodiment, Figure 6 As shown, a computing power provider determination device is provided, comprising: an acquisition module 11, a reward calculation module 12 and a computing power provider determination module 13, wherein:

[0139] The acquisition module 11 is used to obtain the computing power demand information of the computing power demander and the computing power resource description information of each computing power provider;

[0140] The reward calculation module 12 is used to determine, for each computing power provider, the provider reward value of the computing power provider and the demander reward value of the computing power demander according to the computing power demand information and the computing power resource description information of the computing power provider; wherein the provider reward value is the reward value obtained after the computing power provider provides the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider;

[0141] The computing power provider determination module 13 is used to select a target provider corresponding to the computing power demander from among the computing power providers based on the provider reward value of each computing power provider and the demander reward value of each computing power demander; wherein the target provider is used to provide computing power resources to the computing power demander.

[0142] In one embodiment, the computing power resource description information includes resource information of available computing power resources under different computing power types, and the resource information of available computing power resources under each computing power type includes the resource amount, deployment location information and unit computing power resource value of the available computing power resources under the computing power type; the reward calculation module 12 is also used to:

[0143] From different computing power types, select the target computing power type that matches the computing power type requirement in the computing power requirement information;

[0144] When it is determined that the amount of available computing resources under the target computing power type meets the resource amount requirement in the computing power requirement information, select the target computing power resources that meet the resource amount requirement from the available computing power resources under the target computing power type;

[0145] The provider reward value of the computing power provider is determined based on the resource quantity, deployment location information and unit computing power resource value of the target computing power resources.

[0146] In one embodiment, the resource information also includes the computing power performance index of each unit computing power resource in the available computing power resources under the computing power type; the reward calculation module 12 is also used to: select the target computing power resource that meets the resource quantity requirement from the available computing power resources under the target computing power type, including:

[0147] Select a candidate computing resource whose computing power performance index meets the computing power performance requirement in the computing power requirement information from each unit computing power resource included in the available computing power resources under the target computing power type;

[0148] From the candidate computing resources, select the target computing resources that meet the resource requirements.

[0149] In one of the embodiments, the resource information also includes a computing power cost indicator of each unit computing power resource in the available computing power resources under the computing power type; the computing power cost indicator of each unit computing power resource includes the total amount of data that the unit computing power resource can process, the computing cost of processing the total amount of data, and the communication cost; the reward calculation module is further used to: determine the computing power processing capacity of the target computing power resource according to the ratio between the resource amount of the target computing power resource and the resource amount of the available computing power resource of the computing power provider;

[0150] The provider reward value of the computing power provider is determined based on the computing power cost indicators, deployment location information and unit computing power resource value of each unit computing power resource in the target computing power resources, as well as the computing power processing capacity of the target computing power resources.

[0151] In one of the embodiments, the reward calculation module 12 is also used for: the computing power provider determines the demander reward value of the computing power demander based on the computing power processing capacity of the target computing power resources, the resource quantity of the target computing power resources and the computing power value of the unit computing power resources in the target computing power resources.

[0152] In one embodiment, the computing power provider determination module 13 is further used to: determine a provider set according to the computing power providers corresponding to the reward values ​​of each demander of the computing power demander;

[0153] The computing power provider with the largest demander reward value in the provider set is selected as the candidate provider;

[0154] determining whether the provider reward value of the candidate provider is greater than or equal to the expected reward value;

[0155] If yes, the candidate provider will be used as the target provider corresponding to the computing power demander;

[0156] If not, the computing power provider corresponding to the largest demander reward value is selected from the providers other than the candidate provider in the provider set as the updated candidate provider, and based on the updated candidate provider, the step of determining whether the provider reward value of the candidate provider is greater than or equal to the expected reward value is returned to execute.

[0157] Each module in the computing power provider determination device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0158] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 7As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a method for determining a computing power provider is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0159] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0160] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0161] Obtain computing power demand information of computing power demanders and computing power resource description information of each computing power provider;

[0162] For each computing power provider, determine the provider reward value of the computing power provider and the demander reward value of the computing power demander based on the computing power demand information and the computing power resource description information of the computing power provider; the provider reward value is the reward value obtained by the computing power provider after providing the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider;

[0163] According to the provider reward value of each computing power provider and the demander reward value of each computing power demander, a target provider corresponding to the computing power demander is selected from each computing power provider; wherein the target provider is used to provide computing power resources to the computing power demander.

[0164] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0165] Obtain computing power demand information of computing power demanders and computing power resource description information of each computing power provider;

[0166] For each computing power provider, determine the provider reward value of the computing power provider and the demander reward value of the computing power demander based on the computing power demand information and the computing power resource description information of the computing power provider; the provider reward value is the reward value obtained by the computing power provider after providing the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider;

[0167] According to the provider reward value of each computing power provider and the demander reward value of each computing power demander, a target provider corresponding to the computing power demander is selected from each computing power provider; wherein the target provider is used to provide computing power resources to the computing power demander.

[0168] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0169] Obtain computing power demand information of computing power demanders and computing power resource description information of each computing power provider;

[0170] For each computing power provider, determine the provider reward value of the computing power provider and the demander reward value of the computing power demander based on the computing power demand information and the computing power resource description information of the computing power provider; the provider reward value is the reward value obtained by the computing power provider after providing the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider;

[0171] According to the provider reward value of each computing power provider and the demander reward value of each computing power demander, a target provider corresponding to the computing power demander is selected from each computing power provider; wherein the target provider is used to provide computing power resources to the computing power demander.

[0172] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0173] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0174] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0175] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for determining a computing power provider, characterized in that: The method comprises: Obtain computing power demand information of computing power demanders and computing power resource description information of each computing power provider; For each computing power provider, determine the provider reward value of the computing power provider and the demander reward value of the computing power demander according to the computing power demand information and the computing power resource description information of the computing power provider; wherein the provider reward value is the reward value obtained after the computing power provider provides the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained by the computing power demander after using the computing power resources provided by the computing power provider; According to the provider reward value of each computing power provider and the demander reward value of each computing power demander, a target provider corresponding to the computing power demander is selected from each computing power provider; wherein the target provider is used to provide computing power resources to the computing power demander.

2. The method according to claim 1, characterized in that: The computing power resource description information includes resource information of available computing power resources under different computing power types, and the resource information of available computing power resources under each computing power type includes the resource quantity, deployment location information and unit computing power resource value of the available computing power resources under the computing power type; The determining, according to the computing power demand information and the computing power resource description information of the computing power provider, a provider reward value of the computing power provider includes: Selecting a target computing power type that matches the computing power type requirement in the computing power requirement information from different computing power types; When it is determined that the amount of available computing power resources under the target computing power type meets the resource amount requirement in the computing power requirement information, select a target computing power resource that meets the resource amount requirement from the available computing power resources under the target computing power type; The provider reward value of the computing power provider is determined based on the resource quantity, deployment location information and unit computing power resource value of the target computing power resources.

3. The method according to claim 2, characterized in that The resource information also includes the computing power performance index of each unit computing power resource in the available computing power resources under the computing power type; The selecting a target computing resource that meets the resource quantity requirement from the available computing resources under the target computing type includes: Selecting, from each unit computing power resource included in the available computing power resources under the target computing power type, a candidate computing power resource whose computing power performance indicator meets the computing power performance requirement in the computing power requirement information; From the candidate computing resources, select a target computing resource that meets the resource quantity requirement.

4. The method according to claim 2, characterized in that: The resource information also includes a computing power cost indicator of each unit computing power resource in the available computing power resources under the computing power type; the computing power cost indicator of each unit computing power resource includes the total amount of data that can be processed by the unit computing power resource, the computing cost and the communication cost of processing the total amount of data; The determining the provider reward value of the computing power provider according to the resource amount, deployment location information and unit computing power resource value of the target computing power resource includes: Determine the computing power processing capacity of the target computing power resource according to the ratio between the resource amount of the target computing power resource and the resource amount of the available computing power resource of the computing power provider; The provider reward value of the computing power provider is determined based on the computing power cost index, deployment location information and unit computing power resource value of each unit computing power resource in the target computing power resource, and the computing power processing capacity of the target computing power resource.

5. The method according to claim 4, characterized in that Determining the demander reward value of the computing power demander according to the computing power demand information and the computing power resource description information of the computing power provider, including: The computing power provider determines the demander reward value of the computing power demander based on the computing power processing capacity of the target computing power resources, the resource volume of the target computing power resources, and the computing power value of the unit computing power resources in the target computing power resources.

6. The method according to claim 1, characterized in that The selecting a target provider corresponding to the computing power demander from each computing power provider according to the provider reward value of each computing power provider and each demander reward value of the computing power demander comprises: Determine a provider set according to the computing power providers corresponding to the reward values ​​of each computing power demander; The computing power provider with the largest reward value from the demander in the provider set is selected as a candidate provider; determining whether the provider reward value of the candidate provider is greater than or equal to the expected reward value; If yes, the candidate provider is used as the target provider corresponding to the computing power demander; If not, then from the providers other than the target provider in the provider set, the computing power provider corresponding to the largest demander reward value is taken as the updated candidate provider, and based on the updated candidate provider, return to execute the step of determining whether the provider reward value of the candidate provider is greater than or equal to the expected reward value.

7. A computing power provider determination device, characterized in that: The device comprises: The acquisition module is used to obtain the computing power demand information of the computing power demander and the computing power resource description information of each computing power provider; A reward calculation module is used to determine, for each computing power provider, a provider reward value of the computing power provider and a demander reward value of the computing power demander according to the computing power demand information and the computing power resource description information of the computing power provider; wherein the provider reward value is the reward value obtained after the computing power provider provides the computing power demander with computing power resources that meet the computing power demand information, and the demander reward value is the reward value obtained after the computing power demander uses the computing power resources provided by the computing power provider; A computing power provider determination module is used to select a target provider corresponding to the computing power demander from among the computing power providers based on the provider reward value of each computing power provider and the demander reward value of each computing power demander; wherein the target provider is used to provide computing power resources to the computing power demander.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images