Resource deployment method and device, computer equipment and storage medium

By using the combination of task allocation model and multiple single-item models in edge computing, we optimize resource configuration, and solve the problems of limited resources and uneven distribution of edge computing resources, improving the efficiency and real-time nature of task processing.

CN119088551APending Publication Date: 2024-12-06GUANGZHOU POWER SUPPLY BUREAU GUANGDONG POWER GRID CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411129829.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In edge computing technology, due to limited resources and uneven distribution, the transmission time of computing tasks is delayed, which cannot meet the real-time requirements, and the resource allocation is uneven, affecting the task processing efficiency.

Method used

Using a pre-built task allocation model, based on the pending task set and the edge server set, the allocation and processing tasks of each edge server are obtained, and resource deployment is carried out based on the edge server information and task information. Through the combination and weight calculation of multiple single models, Optimize task allocation and resource configuration to ensure timeliness and resource uniformity of data transmission.

Benefits of technology

It improves the uniformity of task allocation and resource allocation, reduces the transmission delay of calculation tasks, improves the efficiency of task processing, and meets the requirements of real-time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119088551A_ABST
    Figure CN119088551A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides a resource deployment method and device, computer equipment, a storage medium and a computer program product, and relates to the technical field of cloud computing. The method comprises the steps of obtaining a to-be-processed task set and an edge server set of a task processing system; the to-be-processed task set comprises to-be-processed task information of each to-be-processed task in the multiple to-be-processed tasks; the edge server set comprises edge server information of each edge server in the plurality of edge servers; by utilizing a pre-constructed task allocation model, based on the to-be-processed task set and the edge server set, acquiring allocation processing tasks of each edge server; and performing resource deployment on each edge server according to the edge server information and the to-be-processed task information of the allocated processing task of each edge server. According to the method, the uniformity of task allocation and resource allocation is improved, and then the task processing efficiency can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

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

[0002] In some power scenarios that have high requirements for data processing capabilities, cloud computing is usually required. Data centers have abundant resources and powerful data processing capabilities, which can achieve optimal configuration and efficient scheduling of power resources, and significantly improve the operating efficiency and reliability of power systems. In general, cloud computing data centers are located at the core of the network. When users transmit computing tasks to cloud computing cloud servers, they need to go through the wireless access network to the core network, and then forward through multiple hops to reach the central cloud. This transmission method makes delay-sensitive application services respond slowly and cannot meet real-time requirements. In order to solve this problem, edge computing technology can be used to provide cloud computing capabilities at the edge of the network by approaching the user side, and migrate the network's functions, content, and resources to locations closer to end users to support computing-intensive network services, thereby reducing the transmission delay of computing tasks and the computing load of remote cloud servers.

[0003] However, in the current edge computing technology, due to the limited resources of edge computing, it will lead to insufficient resources and uneven distribution of resources on multiple edge servers. Summary of the invention

[0004] Based on this, it is necessary to provide a resource deployment method, apparatus, computer equipment, storage medium and computer program product to address the above technical issues.

[0005] In a first aspect, the present application provides a resource deployment method. The method comprises:

[0006] Acquire a set of pending tasks and an edge server set of a task processing system; the set of pending tasks includes pending task information of each of the plurality of pending tasks; the edge server set includes edge server information of each of the plurality of edge servers;

[0007] Using a pre-built task allocation model, based on the set of pending tasks and the set of edge servers, the allocated processing tasks of each edge server are obtained; the allocated processing tasks are pending tasks allocated to each edge server;

[0008] Resource deployment is performed on each edge server according to the edge server information and the to-be-processed task information of the assigned processing tasks of each edge server.

[0009] In one embodiment, the task allocation model includes a first single model and a second single model; the use of the pre-built task allocation model to obtain the allocated processing tasks of each edge server based on the set of tasks to be processed and the set of edge servers includes: using the first single model to obtain the first allocated subtask of each edge server based on the set of tasks to be processed and the set of edge servers; using the second single model to obtain the second allocated subtask of each edge server based on the set of tasks to be processed and the set of edge servers; and obtaining the allocated processing tasks of each edge server based on the first allocated subtask and the second allocated subtask.

[0010] In one embodiment, obtaining the allocation processing task of each edge server according to the first allocation subtask and the second allocation subtask includes: determining a first weight value of the first single item model; determining a second weight value of the second single item model; and obtaining the allocation processing task of each edge server according to the first weight value, the first allocation subtask, the second weight value and the second allocation subtask.

[0011] In one of the embodiments, the number of the assigned processing tasks is multiple; the resource deployment for each edge server according to the edge server information and the pending task information of the assigned processing tasks of each edge server includes: according to the edge server information and the pending task information, resource deployment for each of the assigned processing tasks is performed in the priority order of the multiple assigned processing tasks.

[0012] In one embodiment, the performing of resource deployment for each of the assigned processing tasks includes:

[0013] According to the edge server information and the task information to be processed, a delay threshold is determined for each of the assigned processing tasks; if each of the assigned processing tasks satisfies the delay threshold determination, the edge server information of the corresponding edge server is updated.

[0014] In one of the embodiments, the edge server information includes the maximum available resources and the currently available resources of the corresponding edge server; the pending task information includes the resource consumption of the corresponding assigned processing task; the delay threshold determination is performed on each of the assigned processing tasks based on the edge server information and the pending task information, including: determining the difference between the currently available resources and the resource consumption as the remaining currently available resources corresponding to the edge server; and performing the delay threshold determination on each of the assigned processing tasks based on the ratio of the remaining currently available resources to the resource consumption.

[0015] In a second aspect, the present application provides a resource deployment device. The device comprises:

[0016] An acquisition module, used to acquire a set of pending tasks and an edge server set of a task processing system; the set of pending tasks includes pending task information of each of the plurality of pending tasks; the edge server set includes edge server information of each of the plurality of edge servers;

[0017] A computing module, configured to obtain, by using a pre-built task allocation model, the allocated processing tasks of each edge server based on the set of pending tasks and the set of edge servers; the allocated processing tasks are pending tasks allocated to each edge server;

[0018] The resource deployment module is used to deploy resources to each edge server according to the edge server information and the to-be-processed task information of the allocated processing tasks of each edge server.

[0019] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the following steps when executing the computer program:

[0020] Acquire a set of pending tasks and an edge server set of a task processing system; the set of pending tasks includes pending task information of each of the plurality of pending tasks; the edge server set includes edge server information of each of the plurality of edge servers;

[0021] Using a pre-built task allocation model, based on the set of pending tasks and the set of edge servers, the allocated processing tasks of each edge server are obtained; the allocated processing tasks are pending tasks allocated to each edge server;

[0022] Resource deployment is performed on each edge server according to the edge server information and the to-be-processed task information of the assigned processing tasks of each edge server.

[0023] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0024] Acquire a set of pending tasks and an edge server set of a task processing system; the set of pending tasks includes pending task information of each of the plurality of pending tasks; the edge server set includes edge server information of each of the plurality of edge servers;

[0025] Using a pre-built task allocation model, based on the set of pending tasks and the set of edge servers, the allocated processing tasks of each edge server are obtained; the allocated processing tasks are pending tasks allocated to each edge server;

[0026] Resource deployment is performed on each edge server according to the edge server information and the to-be-processed task information of the assigned processing tasks of each edge server.

[0027] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the following steps are implemented:

[0028] Acquire a set of pending tasks and an edge server set of a task processing system; the set of pending tasks includes pending task information of each of the plurality of pending tasks; the edge server set includes edge server information of each of the plurality of edge servers;

[0029] Using a pre-built task allocation model, based on the set of pending tasks and the set of edge servers, the allocated processing tasks of each edge server are obtained; the allocated processing tasks are pending tasks allocated to each edge server;

[0030] Resource deployment is performed on each edge server according to the edge server information and the to-be-processed task information of the assigned processing tasks of each edge server.

[0031] In the above-mentioned resource deployment method, device, computer equipment, storage medium and computer program product, a set of pending tasks and a set of edge servers of a task processing system are obtained; the set of pending tasks includes the pending task information of each pending task among multiple pending tasks; the set of edge servers includes the edge server information of each edge server among multiple edge servers; using a pre-constructed task allocation model, based on the set of pending tasks and the set of edge servers, the assigned processing tasks of each edge server are obtained; the assigned processing tasks are the pending tasks assigned to each edge server; and resources are deployed for each edge server according to the edge server information and the pending task information of the assigned processing tasks of each edge server. In the method provided in the embodiment of the present application, a task allocation model can be introduced, and tasks and resource configuration can be performed on multiple edge servers of the task processing system based on the set of pending tasks and the set of edge servers, which can improve the uniformity of task allocation and resource configuration while ensuring the timeliness of data transmission and task processing, thereby improving the efficiency of task processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] 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.

[0033] Figure 1 An application environment diagram of a resource deployment method provided in an embodiment of the present application;

[0034] Figure 2 A schematic diagram of a resource deployment method provided in an embodiment of the present application;

[0035] Figure 3 A flowchart of obtaining the allocation processing tasks of each edge server using the prediction results of multiple single models provided in an embodiment of the present application;

[0036] Figure 4 A schematic diagram of a flow chart of obtaining the distribution processing tasks of each edge server according to the weight of each single model provided in an embodiment of the present application;

[0037] Figure 5 A structural block diagram of a resource deployment device provided in an embodiment of the present application;

[0038] Figure 6 An internal structure diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0039] 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.

[0040] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0041] 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. The acquisition, storage, use, and processing of data in the technical solution of this application are in compliance with relevant provisions of national laws and regulations.

[0042] The resource configuration method provided in the embodiment of the present application can be applied to Figure 1 The application environment shown. The application environment may include a task processing terminal, a task processing server, a cloud server and an edge server. Among them:

[0043] The user can register and log in to the task processing system on the task processing terminal. The user can trigger a login request on the task processing terminal. In response to the login request, the task processing terminal can display a login interface. The task processing terminal can enter a user name and password on the login interface to log in. Among them, the login method for the user to log in to the task processing system may include but is not limited to applications, public accounts, applets, and web pages, etc., and this application does not limit this. If the user logs in successfully, a task processing request for the pending task can be triggered through the task processing interface of the task processing terminal. In response to the task processing request, the task processing terminal can send the task processing request to the task processing server of the task processing system.

[0044] A task processing server, which can obtain a set of pending tasks and an edge server set of a task processing system; the set of pending tasks includes pending task information of each pending task among multiple pending tasks; the edge server set includes edge server information of each edge server among multiple edge servers; next, a pre-built task allocation model can be used to obtain the allocated processing tasks of each edge server based on the set of pending tasks and the set of edge servers; the allocated processing tasks are the pending tasks allocated to each edge server; finally, resources are deployed to each edge server according to the edge server information and the pending task information of the allocated processing tasks of each edge server; the task resources corresponding to each edge server are sent to the cloud server; the cloud server deploys the task resources corresponding to each edge server to the corresponding edge server.

[0045] In an exemplary embodiment, Figure 2As shown, a resource deployment method is provided. This embodiment uses the method applied to a server as an example for illustration. It can be understood that the method can also be applied to a terminal, and 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. In this embodiment, the method includes the following steps:

[0046] Step 202: Obtain a set of tasks to be processed and a set of edge servers of the task processing system.

[0047] The task processing system may be a cloud-edge collaboration system, which may include a task processing terminal, a task processing server, a cloud server, and an edge server. The task processing system may process pending tasks corresponding to task processing requests triggered by users on the task processing terminal. The pending task set may include multiple pending tasks, and pending task information of each pending task in the multiple pending tasks; the pending task information may include computing resources and storage resources required to execute the corresponding pending tasks, wherein: represents a set of pending tasks, each of which can be represented as ,in Indicates the task The computational resources required for execution are quantified in terms of the CPU power required for each task (GHz / task). Indicates the task The storage resources required for execution. is the task type, Indicates the task For time-sensitive tasks, Indicates the task The edge server set includes multiple edge servers and edge server information of each edge server in the multiple edge servers, and the edge server information may include the maximum available resources and current available resources of the corresponding edge server, the maximum available resources may include the maximum computing resources and the maximum storage resources, and the current available resources may include the current computing resources and the current storage resources. Represents the edge server set, represented by edge servers, represented as Each edge server is represented as ,in represents the computing resources of the edge server, Represents the storage resources of the edge server.

[0048] Step 204: using the pre-built task allocation model, based on the set of tasks to be processed and the set of edge servers, obtain the allocated processing tasks of each edge server.

[0049] Among them, the assigned processing tasks are pending tasks assigned to each edge server. In this step, a pre-built task allocation model can be used. The task allocation model can predict the pending tasks that each edge server in the multiple edge servers can carry, that is, the assigned processing tasks. The task allocation model can be a combined model, and the combined model can include multiple single models. In a possible implementation, the combined model can include a first single model and a second single model. The first single model can be a Prophet model, and the second single model can be a long short-term memory network model (Long Short-Term Memory, LSTM). The calculation formula of the combined model can refer to formula (1):

[0050]

[0051] in, Indicates time A predicted sequence value of the combined model, where the predicted sequence value may include a task prediction value of each edge server in the plurality of edge servers, which may be understood as corresponding task information of the assigned processing task; Indicates the number of single-item models participating in the combination, express Moment The predicted sequence value of the model, Represents the weight of the single model, and the sum of the weights is 1, and In the process of using the combined model to predict the tasks to be processed that each edge server in the multiple edge servers can carry, that is, the process of allocating the processing tasks, each of the multiple single models can be used to obtain the prediction results of each edge server in the multiple edge servers, that is, the distribution of processing tasks. Then, the weights corresponding to each single model can be used to perform weighted summation on the multiple prediction results of the multiple single models to obtain the final prediction result. In one possible method, the weights of each single model can be obtained by the following method, and the weights can be determined by the inverse variance method with smaller error: The value of: The first step is to determine the sum of squared errors of each single model. The sum of squared errors can be obtained by referring to formula (2):

[0052]

[0053] in, For the The sum of squared errors of the models; is the predicted sequence value of each edge server, is the actual value of each edge server.

[0054] Furthermore, the weight of each single model can be determined according to the sum of squared errors of each single model, as shown in formula (3):

[0055]

[0056] In a possible implementation, the training process of the combined model can refer to the following steps:

[0057] 1. Obtain the historical task set and historical edge server set carried by the task processing system in the historical time period corresponding to the current time period;

[0058] 2. Divide the historical task set and historical edge server set into two sets of training sets and test sets. The training set is used to train the model, and the test set is used to test the accuracy of the model;

[0059] 3. The optimization goal of the combined model can be determined as maximizing the task hit rate of the edge server.

[0060] The task hit rate is defined as the proportion of user tasks executed by local edge nodes in the total tasks during the system offloading decision and task processing process. The optimization objective can be expressed as formula (4):

[0061]

[0062] in, is the task type of the task to be processed, which can include delay-sensitive tasks and non-delay-sensitive tasks; The computing resources occupied by edge server tasks do not exceed the maximum computing resources. The storage resources occupied by edge server tasks cannot exceed the maximum storage resources.

[0063] Step 206: Deploy resources to each edge server according to the edge server information and the to-be-processed task information of the assigned processing tasks of each edge server.

[0064] After obtaining the assigned processing tasks of each edge server by using the task allocation model, resources can be deployed for each edge server according to the edge server information and the pending task information of the assigned processing tasks of each edge server. In this step, first, the maximum available resources and the currently available resources of each edge server can be obtained according to the edge server information of each edge server. The maximum available resources may include the maximum computing resources and the maximum storage resources, and the currently available resources may include the current computing resources and the current storage resources; next, the number of assigned processing tasks of each edge server may be multiple, and the pending task information of each assigned processing task in the multiple assigned processing tasks of each edge server may also be referred to as obtaining the task type of the corresponding assigned processing task from the assigned processing task information. The task type may include a first task type and a second task type. The first task type may include time-sensitive and non-time-sensitive tasks. The assigned processing tasks belonging to the first task type may be further classified to obtain multiple types of tasks, and the time-sensitive assigned processing tasks may be classified into time-sensitive and non-time-sensitive tasks. The tasks may include multiple tasks, and the non-time-sensitive allocation and processing tasks may also include multiple tasks, namely, the second task type; and the multiple allocation and processing tasks of each edge server are clustered to obtain the number of tasks corresponding to each task type. Further, the multiple allocation and processing tasks of each edge server can be prioritized according to the task type and the number of tasks corresponding to each task type. In a possible implementation method, for each edge server, the multiple allocation and processing tasks of each edge server can be prioritized according to the principle that the priority of time-sensitive tasks is higher than that of non-time-sensitive tasks, the priority of tasks with a smaller data volume in the same sensitive task is higher than that of tasks with a larger data volume, and the corresponding predicted different types of time-sensitive tasks and different types of non-time-sensitive tasks are arranged in descending order according to the number of tasks. Furthermore, resource deployment for each allocated processing task can be performed according to the priority order of the multiple allocated processing tasks. Specifically, a time-sensitive allocated processing task and a non-time-sensitive allocated processing task are used as examples for explanation. When the task type of the current allocated processing task is time-sensitive, the delay threshold of the current allocated processing task can be determined according to the edge server information and the task to be processed information, as shown in formula (5):

[0065]

[0066] in, is the current computing resource of the corresponding edge server; is the current storage resource of the corresponding edge server; is the maximum computing resource of the corresponding edge server; is the maximum storage resource of the corresponding edge server; Assign the jth processing task to the edge server, i.e., the computing resources currently assigned to the processing task; Allocate storage resources for the jth processing task of the edge server; is the delay threshold.

[0067] If the currently assigned processing task meets the delay threshold, the currently available resources of the corresponding edge server are updated, as shown in formula (6):

[0068]

[0069] in, is the current computing resource of the corresponding edge server; is the current storage resource of the corresponding edge server; Allocate computing resources for processing tasks for the jth edge server; Allocate storage resources for the jth processing task of the edge server.

[0070] Furthermore, it is possible to determine whether the corresponding edge server still has remaining available resources, as shown in formula (7):

[0071]

[0072] in, is the current computing resource of the corresponding edge server; is the current storage resource of the corresponding edge server; is the maximum computing resource of the corresponding edge server; is the maximum storage resource of the corresponding edge server; is the delay threshold.

[0073] When there are remaining available resources on the corresponding edge server, resources can be deployed for the next allocated processing task according to the priority order of the multiple allocated processing tasks; when there are no remaining available resources on the corresponding edge server, the number of allocated processing tasks of the corresponding edge server can be adjusted, and the task configuration and resource configuration can be updated.

[0074] When the task type of the currently assigned processing task is not time-sensitive, the delay threshold of the currently assigned processing task can be determined based on the edge server information and the task to be processed information, as shown in formula (5):

[0075]

[0076] in, is the current computing resource of the corresponding edge server; is the current storage resource of the corresponding edge server; is the maximum computing resource of the corresponding edge server; is the maximum storage resource of the corresponding edge server; Allocate computing resources for processing tasks for the jth edge server; Allocate storage resources for the jth processing task of the edge server; is the delay threshold.

[0077] If the currently assigned processing task meets the delay threshold, the currently available resources of the corresponding edge server are updated, as shown in formula (6):

[0078]

[0079] in, is the current computing resource of the corresponding edge server; is the current storage resource of the corresponding edge server; Allocate computing resources for processing tasks for the jth edge server; Allocate storage resources for the jth processing task of the edge server.

[0080] Furthermore, it is possible to determine whether the corresponding edge server still has remaining available resources, as shown in formula (7):

[0081]

[0082] in, is the current computing resource of the corresponding edge server; is the current storage resource of the corresponding edge server; is the maximum computing resource of the corresponding edge server; is the maximum storage resource of the corresponding edge server; is the delay threshold.

[0083] If there are remaining available resources on the corresponding edge server, resources can be deployed for the next allocated processing task according to the priority order of multiple allocated processing tasks; if there are no remaining available resources on the corresponding edge server, the number of allocated processing tasks on the corresponding edge server can be adjusted, and the task configuration and resource configuration can be updated. The resource deployment process for the corresponding edge server is continued until all the allocated processing tasks of the corresponding edge server are completed. Furthermore, the corresponding resource deployment can be sent to the cloud server, and the cloud server can deploy the task resources corresponding to the resource deployment to the corresponding edge server.

[0084] In the method of this embodiment, a set of pending tasks and a set of edge servers of a task processing system are obtained; the set of pending tasks includes pending task information of each pending task among multiple pending tasks; the set of edge servers includes edge server information of each edge server among multiple edge servers; using a pre-built task allocation model, based on the set of pending tasks and the set of edge servers, the assigned processing tasks of each edge server are obtained; the assigned processing tasks are pending tasks assigned to each edge server; and resources are deployed for each edge server according to the edge server information and the pending task information of the assigned processing tasks of each edge server. In the method provided by the embodiment of the present application, a task allocation model can be introduced, and based on the set of pending tasks and the set of edge servers, tasks can be allocated and resources can be configured for multiple edge servers of the task processing system, which can improve the uniformity of task allocation and resource configuration while ensuring the timeliness of data transmission and task processing, thereby improving the efficiency of task processing.

[0085] In an exemplary embodiment, Figure 3 As shown, step 204 may include steps 302 to 306. Among them:

[0086] Step 302: using the first single item model, based on the set of tasks to be processed and the set of edge servers, obtain a first allocated subtask for each edge server.

[0087] Step 304: using the second single item model, based on the set of tasks to be processed and the set of edge servers, obtain a second allocated subtask for each edge server.

[0088] Step 306: Obtain the allocated processing task of each edge server according to the first allocated subtask and the second allocated subtask.

[0089] The task allocation model includes a first single model and a second single model. In the process of using the combined model to predict the tasks to be processed that each edge server in the multiple edge servers can carry, that is, the process of allocating the processing tasks, each of the multiple single models can be used to obtain the prediction results of each edge server in the multiple edge servers, that is, the allocated processing tasks. Then, the weights corresponding to each single model can be used to perform weighted summation on the multiple prediction results of the multiple single models to obtain the final prediction result, which can be seen in formula (8):

[0090]

[0091] Among them, the For the moment , the prediction result of the corresponding edge server obtained by the task allocation model, that is, the allocation processing task; In order to utilize the first single item model, based on the set of tasks to be processed and the set of edge servers, a first allocated subtask of each edge server is obtained; the first allocated subtask may be an allocated processing task of the corresponding edge server predicted by the first single item model; In order to utilize the second single item model, based on the set of tasks to be processed and the set of edge servers, a second allocated subtask of each edge server is obtained; the second allocated subtask may be an allocated processing task of the corresponding edge server predicted by the second single item model; is the weight corresponding to the first single item model; is the weight corresponding to the second single model.

[0092] In an exemplary embodiment, Figure 4 As shown, step 306 may include steps 402 to 406. Among them:

[0093] Step 402, determining a first weight value of a first single item model.

[0094] Step 404, determining a second weight value of the second single item model.

[0095] Step 406: Obtain the allocated processing task of each edge server according to the first weight value, the first allocated subtask, the second weight value, and the second allocated subtask.

[0096] The weight of each individual model can be determined based on factors such as performance indicators, reliability, and applicability of each individual model. In this exemplary embodiment, the weight of each individual model can be obtained by the following method: The value of: The first step is to determine the sum of squared errors of each single model. The sum of squared errors can be obtained by referring to formula (2):

[0097]

[0098] in, For the The sum of squared errors of the models; is the predicted sequence value of each edge server, is the actual value of each edge server.

[0099] Furthermore, the weight of each single model can be determined according to the sum of squared errors of each single model, as shown in formula (3):

[0100]

[0101] In the method of this embodiment, a combined model can be constructed through multiple single models to predict the task volume of each edge server, thereby avoiding the disadvantage that a single model prediction is often difficult to accurately capture the dynamic changes in the number of tasks, improving the accuracy of task volume prediction, and further improving the accuracy and efficiency of task deployment and resource deployment.

[0102] In an exemplary embodiment, step 206 may include:

[0103] According to the edge server information and the task to be processed information, resources are deployed for each assigned processing task in accordance with the priority order of the multiple assigned processing tasks.

[0104] Specifically, in an exemplary embodiment, it may include: performing a delay threshold determination on each assigned processing task according to the edge server information and the task information to be processed; if each assigned processing task satisfies the delay threshold determination, the edge server information of the corresponding edge server is updated. The delay threshold determination on each assigned processing task may include: determining the difference between the current available resources and the resource consumption as the remaining current available resources corresponding to the edge server; and performing a delay threshold determination on each assigned processing task according to the ratio of the remaining current available resources to the resource consumption.

[0105] First, the maximum available resources and currently available resources of each edge server can be obtained according to the edge server information of each edge server. The maximum available resources may include the maximum computing resources and the maximum storage resources, and the currently available resources may include the current computing resources and the current storage resources. Next, the number of allocated processing tasks of each edge server may be multiple, and the to-be-processed task information of each allocated processing task in the multiple allocated processing tasks of each edge server may also be referred to as obtaining the task type of the corresponding allocated processing task from the allocated processing task information. The task type may include a first task type and a second task type. The first task type may include time-sensitive and non-time-sensitive types. The allocated processing tasks belonging to the first task type may be further classified to obtain multiple types of tasks. Time-sensitive allocated processing tasks may be Including multiple tasks, non-time-sensitive allocation processing tasks can also include multiple tasks, that is, the second task type; and clustering the multiple allocation processing tasks of each edge server to obtain the number of tasks corresponding to each task type, further, the multiple allocation processing tasks of each edge server can be prioritized according to the task type and the number of tasks corresponding to each task type. In a possible implementation, for each edge server, the priority of time-sensitive tasks is higher than that of non-time-sensitive tasks, the priority of tasks with small data volume is higher than that of tasks with large data volume in the same sensitive type, and the corresponding predicted different types of tasks belonging to time-sensitive and different types of tasks belonging to non-time-sensitive are arranged in descending order according to the number of tasks, and the multiple allocation processing tasks of each edge server can be prioritized. Furthermore, the resources for each allocation processing task can be deployed according to the priority order of the multiple allocation processing tasks. Specifically, taking one of the time-sensitive allocation processing tasks as an example, the resource deployment process of the non-time-sensitive allocation processing task is the same as the resource deployment process of the time-sensitive allocation processing task. In the case where the task type of the currently assigned processing task is time-sensitive, the delay threshold of the currently assigned processing task can be determined based on the edge server information and the task information to be processed, as shown in formula (5). If the currently assigned processing task satisfies the delay threshold determination, the currently available resources of the corresponding edge server are updated, as shown in formula (6); further, it can be determined whether the corresponding edge server still has remaining available resources, as shown in formula (7); in the case where the corresponding edge server has remaining available resources, the resources of the next assigned processing task can be deployed according to the priority order of multiple assigned processing tasks; in the case where the corresponding edge server does not have remaining available resources, the number of the assigned processing tasks of the corresponding edge server can be adjusted, and the task configuration and resource configuration can be updated.

[0106] In the method of this embodiment, during the process of resource deployment for each edge server, a delay threshold determination can be performed on each assigned processing task, which can improve the uniformity of task allocation and resource configuration while ensuring the timeliness of data transmission and task processing, thereby improving the efficiency of task processing.

[0107] 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.

[0108] Based on the same inventive concept, the embodiment of the present application also provides a resource deployment device for implementing the resource deployment method involved above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more resource deployment device embodiments provided below can refer to the limitations on the resource deployment method above, and will not be repeated here.

[0109] In one embodiment, Figure 5 As shown, a resource deployment device is provided, including: an acquisition module 502, a calculation module 504 and a resource deployment module 506, wherein:

[0110] An acquisition module 502 is used to acquire a set of pending tasks and an edge server set of a task processing system; the set of pending tasks includes pending task information of each of the plurality of pending tasks; the edge server set includes edge server information of each of the plurality of edge servers;

[0111] The calculation module 504 is used to obtain the assigned processing tasks of each edge server based on the set of pending tasks and the set of edge servers by using a pre-built task allocation model; the assigned processing tasks are pending tasks allocated to each edge server;

[0112] The resource deployment module 506 is used to deploy resources to each edge server according to the edge server information and the to-be-processed task information of the allocated processing tasks of each edge server.

[0113] In one embodiment, the task allocation model includes a first single item model and a second single item model; the calculation module 504 is also used to: use the first single item model to obtain the first allocated subtask of each edge server based on the set of tasks to be processed and the set of edge servers; use the second single item model to obtain the second allocated subtask of each edge server based on the set of tasks to be processed and the set of edge servers; obtain the allocated processing task of each edge server according to the first allocated subtask and the second allocated subtask.

[0114] In one embodiment, the calculation module 504 is also used to: determine the first weight value of the first single item model; determine the second weight value of the second single item model; and obtain the allocation processing task of each edge server based on the first weight value, the first allocation subtask, the second weight value and the second allocation subtask.

[0115] In one embodiment, the resource deployment module 506 is further configured to: perform resource deployment for each of the assigned processing tasks according to the edge server information and the to-be-processed task information and in the priority order of the plurality of assigned processing tasks.

[0116] In one embodiment, the resource deployment module 506 is further used to: perform a delay threshold determination on each of the assigned processing tasks according to the edge server information and the task information to be processed; if each of the assigned processing tasks meets the delay threshold determination, update the edge server information of the corresponding edge server.

[0117] In one embodiment, the edge server information includes the maximum available resources and currently available resources of the corresponding edge server; the to-be-processed task information includes the resource consumption of the corresponding assigned processing task; the resource deployment module 506 is further used to: determine the difference between the currently available resources and the resource consumption as the remaining currently available resources corresponding to the edge server; and perform the delay threshold determination on each of the assigned processing tasks based on the ratio of the remaining currently available resources to the resource consumption.

[0118] Each module in the resource deployment 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.

[0119] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 6As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. 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 resource deployment related data. The network 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 resource deployment method is implemented.

[0120] Those skilled in the art will understand that Figure 6 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.

[0121] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0122] 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 steps in the above-mentioned method embodiments are implemented.

[0123] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0124] 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.

[0125] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods 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 this application can include at least one of non-volatile 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), magnetoresistive 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, etc., but are not limited to this.

[0126] 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 specification.

[0127] 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 resource deployment method, characterized in that: The method comprises: Acquire a set of pending tasks and an edge server set of a task processing system; the set of pending tasks includes pending task information of each of the plurality of pending tasks; the edge server set includes edge server information of each of the plurality of edge servers; Using a pre-built task allocation model, based on the set of pending tasks and the set of edge servers, the allocated processing tasks of each edge server are obtained; the allocated processing tasks are pending tasks allocated to each edge server; Resource deployment is performed on each edge server according to the edge server information and the to-be-processed task information of the assigned processing tasks of each edge server.

2. The method according to claim 1, characterized in that The task allocation model includes a first single item model and a second single item model; The using of the pre-built task allocation model to obtain the allocated processing tasks of each edge server based on the set of tasks to be processed and the set of edge servers includes: Using the first single model, based on the set of tasks to be processed and the set of edge servers, obtaining a first assigned subtask for each edge server; Using the second single model, based on the set of tasks to be processed and the set of edge servers, obtaining a second allocated subtask for each edge server; The allocation processing task of each edge server is acquired according to the first allocation subtask and the second allocation subtask.

3. The method according to claim 2, characterized in that The acquiring the allocated processing task of each edge server according to the first allocated subtask and the second allocated subtask includes: Determining a first weight value of the first single item model; determining a second weight value of the second single item model; The allocated processing task of each edge server is acquired according to the first weight value, the first allocated subtask, the second weight value and the second allocated subtask.

4. The method according to claim 1, characterized in that The number of the assigned processing tasks is multiple; The step of performing resource deployment on each edge server according to the edge server information and the to-be-processed task information of the assigned processing tasks of each edge server includes: According to the edge server information and the to-be-processed task information, resource deployment is performed for each of the assigned processing tasks in accordance with the priority order of the plurality of assigned processing tasks.

5. The method according to claim 1, characterized in that The performing of resource deployment for each of the assigned processing tasks includes: Determine the delay threshold of each of the assigned processing tasks according to the edge server information and the to-be-processed task information; If each of the allocated processing tasks satisfies the delay threshold determination, the edge server information of the corresponding edge server is updated.

6. The method according to claim 5, characterized in that The edge server information includes the maximum available resources and the current available resources of the corresponding edge server; the pending task information includes the resource consumption of the corresponding assigned processing task; The step of determining a delay threshold for each of the assigned processing tasks according to the edge server information and the to-be-processed task information includes: Determine the difference between the currently available resources and the resource consumption as the remaining currently available resources corresponding to the edge server; The delay threshold is determined for each of the allocated processing tasks according to the ratio of the remaining currently available resources to the resource consumption.

7. A resource deployment device, characterized in that: The device comprises: An acquisition module, used to acquire a set of pending tasks and an edge server set of a task processing system; the set of pending tasks includes pending task information of each of the plurality of pending tasks; the edge server set includes edge server information of each of the plurality of edge servers; A computing module, configured to obtain, by using a pre-built task allocation model, the allocated processing tasks of each edge server based on the set of pending tasks and the set of edge servers; the allocated processing tasks are pending tasks allocated to each edge server; The resource deployment module is used to deploy resources to each edge server according to the edge server information and the to-be-processed task information of the allocated processing tasks of each edge server.

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.