Computing power scheduling method and device, computer equipment, readable storage medium and program product

Abstract

The invention relates to a computing power scheduling method and device, computer equipment, a readable storage medium and a program product. The method comprises the following steps: in response to a target computing power demand initiated by a user identifier, converting the target computing power demand into a first vector; performing vector retrieval on the computing power vector database based on the first vector, and determining a target computing power resource from the computing power vector database; and providing the computing power service of the target computing power resource to equipment corresponding to the user identifier. By adopting the method, the cost of computing power scheduling can be reduced.

Description

Technical Field

[0001] The present application relates to the field of cloud computing technology, and in particular to a computing power scheduling method, apparatus, computer equipment, computer-readable storage medium, and computer program product. Background Art

[0002] In the computing power scheduling scenario, in order to match user computing power requirements with available computing power resources, the computing power scheduling platform usually needs to rely on manual operations to process, manually select the computing power resources required by users and provide them to users.

[0003] However, this type of traditional computing power scheduling method has the problem of high cost. Summary of the invention

[0004] Based on this, it is necessary to provide a computing power scheduling method, device, computer equipment, computer-readable storage medium and computer program product that can reduce costs in response to the above technical problems.

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

[0006] In response to a target computing power requirement initiated by a user identifier, convert the target computing power requirement into a first vector;

[0007] Performing a vector search on a computing power vector database based on the first vector, and determining a target computing power resource from the computing power vector database;

[0008] Provide computing power services of the target computing power resources to the device corresponding to the user identifier.

[0009] In one of the embodiments, in response to the target computing power requirement initiated by the user identifier, converting the target computing power requirement into a first vector includes:

[0010] Obtain the target computing power demand initiated by the user ID;

[0011] When the target computing power requirement represents a requirement for new computing power resources, converting the target computing power requirement into a first vector;

[0012] When the target computing power requirement represents the requirement for historical computing power resources, the vector of the historical computing power resources is used as the first vector.

[0013] In one embodiment, performing vector retrieval on a computing power vector database based on the first vector to determine a target computing power resource from the computing power vector database includes:

[0014] Performing a vector search on a computing power vector database based on the first vector, and determining at least two first computing power resources matching the first vector from the computing power vector database;

[0015] Business-related data is obtained, and based on the business-related data, a target computing resource is determined from the at least two first computing resources.

[0016] In one embodiment, the performing vector search on the computing power vector database based on the first vector, before determining the target computing power resource from the computing power vector database, includes:

[0017] Obtain computing power data reported by computing power service providers;

[0018] Performing measurement processing on the computing power data to determine the computing power resources corresponding to the computing power data;

[0019] The vector identifier of the computing power resource is stored in the computing power vector database.

[0020] In one embodiment, storing the vector identifier of the computing power resource in a computing power vector database includes:

[0021] Mapping the computing power resources into low-dimensional vector identifiers through feature engineering;

[0022] The vector identifier is stored in a computing power vector database.

[0023] In one of the embodiments, after providing the computing power service of the target computing power resource to the device corresponding to the user identifier, the method further includes:

[0024] Obtain relevant service data generated during the computing service process of the target computing resource;

[0025] Adding the relevant service data to the computing power data corresponding to the target computing power resource to obtain new computing power data corresponding to the target computing power resource, and performing measurement processing on the new computing power data to determine a new target computing power resource;

[0026] The vector identifier of the new target computing power resource is stored in the computing power vector database.

[0027] In a second aspect, the present application also provides a computing power scheduling device, including:

[0028] A vector conversion module, configured to convert the target computing power requirement into a first vector in response to the target computing power requirement initiated by the user identifier;

[0029] a vector retrieval module, configured to perform a vector search on a computing power vector database based on the first vector, and determine a target computing power resource from the computing power vector database;

[0030] A computing power service providing module is used to provide the computing power service of the target computing power resource to the device corresponding to the user identifier.

[0031] 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:

[0032] In response to a target computing power requirement initiated by a user identifier, convert the target computing power requirement into a first vector;

[0033] Performing vector retrieval on the computing power vector database based on the first vector, and determining a target computing power resource from the computing power vector database;

[0034] Provide computing power services of the target computing power resources to the device corresponding to the user identifier.

[0035] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0036] In response to a target computing power requirement initiated by a user identifier, convert the target computing power requirement into a first vector;

[0037] Performing a vector search on a computing power vector database based on the first vector, and determining a target computing power resource from the computing power vector database;

[0038] Provide computing power services of the target computing power resources to the device corresponding to the user identifier.

[0039] 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:

[0040] In response to a target computing power requirement initiated by a user identifier, convert the target computing power requirement into a first vector;

[0041] Performing a vector search on a computing power vector database based on the first vector, and determining a target computing power resource from the computing power vector database;

[0042] Provide computing power services of the target computing power resources to the device corresponding to the user identifier.

[0043] The above-mentioned computing power scheduling method, apparatus, computer equipment, computer-readable storage medium and computer program product convert the target computing power demand into a first vector in response to the target computing power demand initiated by the user identifier. Then, based on the first vector, a vector search is performed on the computing power vector database, and the target computing power resources can be determined from the computing power vector database, and computing power services of the target computing power resources can be provided to the device corresponding to the user identifier. It does not rely on fixed operating rules, reduces operating manpower costs, reduces the cost of computing power scheduling, improves the utilization rate of computing power resources, avoids the waste of computing power resources, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] 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.

[0045] Figure 1 1 is a flow chart of a computing power scheduling method in one embodiment;

[0046] Figure 2 is a schematic diagram of a computing power scheduling system in one embodiment;

[0047] Figure 3 is a structural block diagram of a computing power scheduling device in one embodiment;

[0048] Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0049] 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.

[0050] In one embodiment, Figure 1As shown, a computing power scheduling method is provided. This embodiment uses the method applied to a computer device as an example for illustration. The computer device can be a terminal or a server. It can be understood that the method can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server can be an independent physical server, such as a computing power scheduling server (computing power scheduling platform), or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0051] In this embodiment, the computing power scheduling method includes the following steps:

[0052] Step S102: In response to the target computing power requirement initiated by the user identifier, convert the target computing power requirement into a first vector.

[0053] The target computing power requirement is the computing power resource requirement required by the device where the user ID is located. The first vector is the vector representation of the target computing power requirement, that is, the vector search input.

[0054] Optionally, in response to the target computing power demand initiated by the user identifier, the target computing power demand is converted into a first vector, including: obtaining the target computing power demand initiated by the user identifier; when the target computing power demand represents the demand for new computing power resources, converting the target computing power demand into the first vector; when the target computing power demand represents the demand for historical computing power resources, using the vector of the historical computing power resources as the first vector.

[0055] The target computing power requirement includes the required computing power resources, which may include new computing power resources or historical computing power resources.

[0056] Step S104: perform vector search on the computing power vector database based on the first vector, and determine the target computing power resource from the computing power vector database.

[0057] Among them, the computing power vector database stores vectors of computing power resources.

[0058] Optionally, performing a vector search on a computing power vector database based on the first vector, and determining a target computing power resource from the computing power vector database, includes: performing a vector search on the computing power vector database based on the first vector, and determining at least two first computing power resources matching the first vector from the computing power vector database; obtaining business-related data, and determining the target computing power resource from the at least two first computing power resources based on the business-related data.

[0059] The first computing resource is a computing resource that matches the first vector, and is a computing resource obtained after primary screening. The vector corresponding to the first computing resource has the same spatial dimension as the first vector. Business-related data may include cost price, geographic location, and recommendation strategy.

[0060] Optionally, the computer device performs secondary filtering or reordering on at least two first computing resources based on business-related data to determine the target computing resources.

[0061] It can be understood that vector-based retrieval is essentially searching for data similar to the first vector from massive vector data. The vector retrieval input is a vector representation and needs to have the same dimension as the target vector space. Therefore, the computing power requirement input by the user needs to be converted into the first vector required for vector retrieval.

[0062] Step S106: Provide computing power services of the target computing power resources to the device corresponding to the user identifier.

[0063] Optionally, the computer device initiates a resource activation process to the computing power service provider to which the target computing power resource belongs, so as to provide computing power services of the target computing power resource to the device corresponding to the user identifier.

[0064] Optionally, after providing the computing power service of the target computing power resource to the device corresponding to the user identifier, it also includes: obtaining relevant service data generated in the process of providing the computing power service of the target computing power resource; adding the relevant service data to the computing power data corresponding to the target computing power resource to obtain new computing power data corresponding to the target computing power resource, and measuring the new computing power data to determine the new target computing power resource; storing the vector identifier of the new target computing power resource in the computing power vector database.

[0065] Optionally, during the process of the target computing power resource performing computing power services, the computer device obtains relevant service data generated in the process, and adds the relevant service data as part of the computing power data of the target computing power resource to the computing power data of the target computing power resource to obtain new computing power data of the target computing power resource.

[0066] It is understandable that the computing power service server can continuously optimize the computing power resource vector model by collecting relevant service data from the user's usage process and continuously enriching the computing power resource feature data set. The usage data generated by the user's usage process can also be used as the data set input of the vector model to form a process that can be continuously optimized. For example, by learning the relationship between user characteristics and computing power characteristics, the matching degree of computing power recommendations for users with similar characteristics can be improved.

[0067] The above-mentioned computing power scheduling method converts the target computing power demand into a first vector in response to the target computing power demand initiated by the user identifier. Then, by performing a vector search on the computing power vector database based on the first vector, the target computing power resources can be determined from the computing power vector database, and the computing power service of the target computing power resources can be provided to the device corresponding to the user identifier. It does not rely on fixed operating rules, reduces operating manpower costs, and reduces the cost of computing power scheduling.

[0068] Matching computing power demand and computing power resources through vector retrieval does not rely on fixed operating rules, which can reduce operating labor costs, improve matching accuracy, and reduce the failure rate of computing power demand matching and the idle rate of computing power resources. Introducing user usage indicators (related service data) as the characteristic components of computing power resources can enable the computing power scheduling platform to have the ability to continuously learn, and obtain more accurate computing power recommendation performance as user usage increases.

[0069] In one embodiment, a vector search is performed on a computing power vector database based on a first vector, before determining a target computing power resource from the computing power vector database, including: obtaining computing power data reported by a computing power service provider; performing measurement processing on the computing power data to determine the computing power resource corresponding to the computing power data; and storing the vector identifier of the computing power resource in the computing power vector database.

[0070] Computing power data includes standard indicators collected from computing power measurements, differentiated indicators reported by service providers, and label indicators generated by operations.

[0071] Optionally, the computing power service provider accesses the computer device (computing power scheduling server) and reports the computing power data through the platform interface according to its own data opening strategy; the computer device obtains the computing power data reported by the computing power service provider.

[0072] Optionally, the computer device (computing power scheduling server) includes a computing power data service, through which the computing power data reported by the computing power service provider is associated with the computing power resource identifier and stored locally on the computer device; when it is necessary to obtain computing power data from this time, it can be obtained according to the computing power resource identifier.

[0073] The computer equipment performs measurement processing on the computing power data, collects the measurement data and stores it locally, and determines the computing power resources corresponding to the computing power data based on the measurement data.

[0074] The computer device obtains the associated tag of the computing power resource. When performing vector retrieval on the computing power vector database, the target computing power resource can be more accurately determined from the computing power vector database based on the associated tag.

[0075] It is understandable that platform operators associate tags with the connected computing resources according to the platform operation tag management system. The associated tags of computing resources, such as characteristic tags such as "high performance computing" and "IO optimization", or application scenario tags such as "computing intensive" and "IO intensive", depend on the tag management system of platform operation. Operation tags are an artificial supplement to the description of computing resource characteristics, which helps to enrich the feature dimensions and thus better distinguish and match computing resources.

[0076] Optionally, storing the vector identifier of the computing power resource in a computing power vector database includes: mapping the computing power resource into a low-dimensional vector identifier through feature engineering; and storing the vector identifier in the computing power vector database.

[0077] Feature engineering is a process that converts data into features that better represent potential problems, thereby improving machine learning Q performance. Feature engineering is the process of converting raw data into features that can describe the data well, and the performance of the model built using them on unknown data can reach the best (or close to the best performance).

[0078] Optionally, the computer device obtains various data of computing power resources, merges the various data of computing power resources into a data set, and maps the data set of computing power resources into a low-dimensional vector identifier through feature engineering.

[0079] It is understandable that as the size of the data set grows, new data on computing resources can be obtained periodically and repeatedly to continuously optimize the data distribution in the vector space. The algorithms and tools used in the dimensionality reduction process will also be adjusted according to changes in data characteristics to improve the accuracy of the model. Through dimensionality reduction technology, it is mapped to a suitable feature vector space, and similarity vector retrieval is used to match user needs when scheduling computing resources.

[0080] In this embodiment, the computer device obtains the computing power data reported by the computing power service provider, and performs measurement processing on the computing power data, so as to accurately determine the computing power resources corresponding to the computing power data, and store the vector identifier of the computing power resources in the computing power vector database, so that the target computing power resources can be determined more accurately from the computing power vector database during computing power scheduling.

[0081] Through data dimensionality reduction, high-dimensional computing data is mapped to low-dimensional vectors, which is adaptive and can retain the characteristics of computing resources as much as possible while improving retrieval performance and greatly reducing the labor cost of docking data and application indicators. Manual data docking requires manual setting of rules or development of system logic for each dimension of data, and the system and data are deeply coupled. In contrast, dimensionality reduction can automatically extract features from the data, and there is no need to adjust the system architecture when the dimension increases or changes. The system and data are relatively decoupled, so it is considered to be more adaptive.

[0082] In one embodiment, Figure 2 As shown, a computer device is taken as a computing power scheduling server (computing power scheduling platform) as an example for explanation. The computing power scheduling server includes computing power scheduling services, computing power data services and computing power retrieval services.

[0083] The computing power data service is used to manage various raw data sets related to computing power resources, including but not limited to the following:

[0084] 1. Computing power indicators collected uniformly by the computing power measurement of the platform, such as CPU computing rate, memory throughput bandwidth, GPU floating-point computing rate, etc. This type of data is defined and managed by the computing power network platform and is universal;

[0085] 2. Computing power data provided by the connected computing power service provider, such as detailed hardware parameters, training speed of specific models, etc. This type of data is defined and reported by the service provider and varies from one service provider to another;

[0086] 3. The computing power labels associated with computing power resources by platform operators in the management background. These labels are usually classified text labels.

[0087] 4. Other data that helps characterize computing resources, such as user usage data;

[0088] 5. Responsible for connecting to obtain and store the required raw data, and providing a data access interface to the computing power retrieval service.

[0089] The computing power retrieval service is used to vectorize the features of the computing power resource data set. Specifically, it can use mainstream methods in the industry such as feature engineering or Embedding technology to map high-dimensional data to low-dimensional vector space while retaining feature information, and store the vectorized representation of computing power resources in a database that supports vector types or a dedicated vector database. The computing power retrieval service provides a vector conversion interface for converting input data into corresponding vector representations in the feature vector space through the same feature function or Embedding model. The computing power retrieval service also provides a vector similarity retrieval interface for matching several similar computing power resource feature vectors for the input vector. Specifically, it can use mainstream vector similarity algorithms in the industry such as Euclidean distance or Manhattan distance.

[0090] The computing power scheduling service is used to process user computing power requests, call the computing power retrieval service to obtain several candidate computing power resources that best match user needs, and process resource activation requests after user confirmation.

[0091] The computing power service provider accesses the computing power scheduling server and reports the computing power data to the computing power scheduling server; the computing power scheduling server connects to the computing power data reported by the computing power service provider through the service provider reporting indicator module, measures the computing power data through the computing power measurement indicator module, determines the computing power resources corresponding to the computing power data, and stores the vector identifier of the computing power resources in the computing power vector database. The operator can associate labels with the vector identifiers of the computing power resources in the computing power vector database through the platform operation label module. The user initiates the target computing power demand through the user usage data module, and the computing power scheduling service responds to the target computing power demand initiated by the user identifier, converts the target computing power demand into a first vector, and performs a similarity vector search on the computing power vector database (computing power feature space) based on the first vector through the computing power retrieval service, determines the target computing power resources from the computing power vector database, and then provides computing power services of the target computing power resources to the corresponding device of the user.

[0092] In this embodiment, computing power resources can be more fully characterized, and as much computing power data and indicators as possible can be adopted; the adaptability is improved, and it does not rely on fixed rules. When facing the growth of data volume and dimensional expansion, the appropriate feature vector space can be adaptively adjusted; cost-effectiveness is improved, and dependence on human management and operation is reduced; the scheduling and matching failure rate is reduced, and the utilization rate of computing power resources is improved. In general, this embodiment can realize the efficient management and scheduling of computing power resources, can better match the computing power needs of users, and has broad application prospects and market potential.

[0093] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned 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 does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can 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 to be 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.

[0094] Based on the same inventive concept, the embodiment of the present application also provides a computing power scheduling device for implementing the computing power scheduling 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 scheduling device embodiments provided below can refer to the limitations of the computing power scheduling method above, and will not be repeated here.

[0095] In an exemplary embodiment, Figure 3 As shown, a computing power scheduling device is provided, including: a vector conversion module 302, a vector retrieval module 304 and a computing power service providing module 306, wherein:

[0096] The vector conversion module 302 is used to convert the target computing power requirement into a first vector in response to the target computing power requirement initiated by the user identifier.

[0097] The vector retrieval module 304 is used to perform a vector search on the computing power vector database based on the first vector, and determine the target computing power resource from the computing power vector database.

[0098] The computing power service providing module 306 is used to provide computing power services of target computing power resources to the device corresponding to the user identifier.

[0099] The above-mentioned computing power scheduling method, apparatus, computer equipment, computer-readable storage medium and computer program product convert the target computing power demand into a first vector in response to the target computing power demand initiated by the user identifier. Then, based on the first vector, a vector search is performed on the computing power vector database, and the target computing power resources can be determined from the computing power vector database, and computing power services of the target computing power resources can be provided to the device corresponding to the user identifier. It does not rely on fixed operating rules, reduces operating manpower costs, and reduces the cost of computing power scheduling.

[0100] In one embodiment, the above-mentioned vector conversion module 302 is also used to obtain the target computing power demand initiated by the user identifier; when the target computing power demand represents the demand for new computing power resources, the target computing power demand is converted into a first vector; when the target computing power demand represents the demand for historical computing power resources, the vector of the historical computing power resources is used as the first vector.

[0101] In one embodiment, the vector retrieval module 304 is further used to perform vector retrieval on a computing power vector database based on the first vector, determine at least two first computing power resources matching the first vector from the computing power vector database; obtain business-related data, and determine a target computing power resource from the at least two first computing power resources based on the business-related data.

[0102] In one embodiment, the above-mentioned device also includes a computing power measurement module; the computing power measurement module is used to obtain computing power data reported by a computing power service provider; perform measurement processing on the computing power data to determine the computing power resources corresponding to the computing power data; and store the vector identifier of the computing power resource in a computing power vector database.

[0103] In one embodiment, the computing power measurement module is also used to map computing power resources into low-dimensional vector identifiers through feature engineering; and store the vector identifiers in a computing power vector database.

[0104] In one embodiment, the computing power measurement module is also used to obtain relevant service data generated during the computing power service process of the target computing power resource; add the relevant service data to the computing power data corresponding to the target computing power resource, obtain new computing power data corresponding to the target computing power resource, and measure the new computing power data to determine the new target computing power resource; store the vector identifier of the new target computing power resource in the computing power vector database.

[0105] Each module in the above computing power scheduling 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.

[0106] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 4 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, 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, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as vector identifiers and computing power data of computing power resources. 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 through a network connection. When the computer program is executed by the processor, a computing power scheduling method is implemented.

[0107] Those skilled in the art will understand that Figure 4The 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.

[0108] 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 the processor implements the following steps when executing the computer program: in response to a target computing power demand initiated by a user identifier, converting the target computing power demand into a first vector; performing vector retrieval on a computing power vector database based on the first vector, and determining the target computing power resources from the computing power vector database; and providing computing power services of the target computing power resources to a device corresponding to the user identifier.

[0109] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the target computing power demand initiated by the user identifier; when the target computing power demand represents the demand for new computing power resources, converting the target computing power demand into a first vector; when the target computing power demand represents the demand for historical computing power resources, using the vector of the historical computing power resources as the first vector.

[0110] In one embodiment, when the processor executes the computer program, the following steps are also implemented: performing a vector search on a computing power vector database based on the first vector, and determining at least two first computing power resources matching the first vector from the computing power vector database; obtaining business-related data, and determining a target computing power resource from the at least two first computing power resources based on the business-related data.

[0111] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining computing power data reported by the computing power service provider; performing measurement processing on the computing power data to determine the computing power resources corresponding to the computing power data; and storing the vector identifier of the computing power resource in a computing power vector database.

[0112] In one embodiment, when the processor executes the computer program, the following steps are also implemented: mapping computing power resources into low-dimensional vector identifiers through feature engineering; and storing the vector identifiers in a computing power vector database.

[0113] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining relevant service data generated during the computing power service process of the target computing power resource; adding the relevant service data to the computing power data corresponding to the target computing power resource to obtain new computing power data corresponding to the target computing power resource, and measuring the new computing power data to determine the new target computing power resource; storing the vector identifier of the new target computing power resource in the computing power vector database.

[0114] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: in response to a target computing power demand initiated by a user identifier, converting the target computing power demand into a first vector; performing a vector search on a computing power vector database based on the first vector, and determining the target computing power resource from the computing power vector database; and providing computing power services of the target computing power resource to a device corresponding to the user identifier.

[0115] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the target computing power demand initiated by the user identifier; when the target computing power demand represents the demand for new computing power resources, converting the target computing power demand into a first vector; when the target computing power demand represents the demand for historical computing power resources, using the vector of historical computing power resources as the first vector.

[0116] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: performing vector retrieval on a computing power vector database based on the first vector, and determining at least two first computing power resources matching the first vector from the computing power vector database; obtaining business-related data, and determining a target computing power resource from the at least two first computing power resources based on the business-related data.

[0117] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining computing power data reported by a computing power service provider; performing measurement processing on the computing power data to determine the computing power resources corresponding to the computing power data; and storing the vector identifier of the computing power resource in a computing power vector database.

[0118] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented: mapping computing power resources into low-dimensional vector identifiers through feature engineering; and storing the vector identifiers in a computing power vector database.

[0119] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining relevant service data generated during the computing power service process of the target computing power resource; adding the relevant service data to the computing power data corresponding to the target computing power resource to obtain new computing power data corresponding to the target computing power resource, and measuring the new computing power data to determine the new target computing power resource; storing the vector identifier of the new target computing power resource in the computing power vector database.

[0120] In one embodiment, a computer program product is provided, including a computer program, which, when executed by a processor, implements the following steps: in response to a target computing power demand initiated by a user identifier, converting the target computing power demand into a first vector; performing a vector search on a computing power vector database based on the first vector, and determining a target computing power resource from the computing power vector database; and providing computing power services of the target computing power resources to a device corresponding to the user identifier.

[0121] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining the target computing power demand initiated by the user identifier; when the target computing power demand represents the demand for new computing power resources, converting the target computing power demand into a first vector; when the target computing power demand represents the demand for historical computing power resources, using the vector of historical computing power resources as the first vector.

[0122] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: performing vector retrieval on a computing power vector database based on the first vector, and determining at least two first computing power resources matching the first vector from the computing power vector database; obtaining business-related data, and determining a target computing power resource from the at least two first computing power resources based on the business-related data.

[0123] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining computing power data reported by a computing power service provider; performing measurement processing on the computing power data to determine the computing power resources corresponding to the computing power data; and storing the vector identifier of the computing power resource in a computing power vector database.

[0124] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented: mapping computing power resources into low-dimensional vector identifiers through feature engineering; and storing the vector identifiers in a computing power vector database.

[0125] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining relevant service data generated during the computing power service process of the target computing power resource; adding the relevant service data to the computing power data corresponding to the target computing power resource to obtain new computing power data corresponding to the target computing power resource, and measuring the new computing power data to determine the new target computing power resource; storing the vector identifier of the new target computing power resource in the computing power vector database.

[0126] 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.

[0127] 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.

[0128] 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.

[0129] 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 computing power scheduling method, characterized in that: The method comprises: In response to a target computing power requirement initiated by a user identifier, convert the target computing power requirement into a first vector; Performing a vector search on a computing power vector database based on the first vector, and determining a target computing power resource from the computing power vector database; Provide computing power services of the target computing power resources to the device corresponding to the user identifier.

2. The method according to claim 1, characterized in that The step of converting the target computing power requirement into a first vector in response to the target computing power requirement initiated by the user identifier includes: Obtain the target computing power demand initiated by the user ID; When the target computing power requirement represents a requirement for new computing power resources, converting the target computing power requirement into a first vector; When the target computing power requirement represents the requirement for historical computing power resources, the vector of the historical computing power resources is used as the first vector.

3. The method according to claim 1, characterized in that The performing a vector search on a computing power vector database based on the first vector to determine a target computing power resource from the computing power vector database includes: Performing a vector search on a computing power vector database based on the first vector, and determining at least two first computing power resources matching the first vector from the computing power vector database; Business-related data is obtained, and based on the business-related data, a target computing resource is determined from the at least two first computing resources.

4. The method according to claim 1, characterized in that: Before performing vector retrieval on the computing power vector database based on the first vector and determining the target computing power resource from the computing power vector database, the method includes: Obtain computing power data reported by computing power service providers; Performing measurement processing on the computing power data to determine the computing power resources corresponding to the computing power data; The vector identifier of the computing power resource is stored in the computing power vector database.

5. The method according to claim 4, characterized in that The storing of the vector identifier of the computing power resource in the computing power vector database includes: Mapping the computing power resources into low-dimensional vector identifiers through feature engineering; The vector identifier is stored in a computing power vector database.

6. The method according to claim 1, characterized in that After providing the computing power service of the target computing power resource to the device corresponding to the user identifier, the method further includes: Obtain relevant service data generated during the computing service process of the target computing resource; Adding the relevant service data to the computing power data corresponding to the target computing power resource to obtain new computing power data corresponding to the target computing power resource, and performing measurement processing on the new computing power data to determine a new target computing power resource; The vector identifier of the new target computing power resource is stored in the computing power vector database.

7. A computing power scheduling device, characterized in that: The device comprises: A vector conversion module, configured to convert the target computing power requirement into a first vector in response to the target computing power requirement initiated by the user identifier; a vector retrieval module, configured to perform a vector search on a computing power vector database based on the first vector, and determine a target computing power resource from the computing power vector database; A computing power service providing module is used to provide the computing power service of the target computing power resource to the device corresponding to the user identifier.

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