A task matching system based on intelligent load balancing and a storage medium

Abstract

The application discloses a kind of task matching system and storage medium based on intelligent load balancing, including deployment n servers, each service deployment m intelligent Agent matcher, establish intelligent Agent matcher matrix;Setting time interval t1, personnel task list loading initialization is carried out every time interval t1, complete personnel task data initialization task;Establish water level feature matrix;Setting time interval t2, calculate water level feature value in water level feature matrix every time interval t2;Setting k task drift intelligent controller.The application provides data basis for big data calculation by Agent matcher matrix to personnel task data, extracts the working state characteristic value of intelligent Agent matcher matrix by establishing water level feature value matrix and water level feature value calculation, improves personnel task matching efficiency by task drift intelligent controller related calculation method, realizes the intelligent load balancing of big data calculation, improves the utilization of computing resources.

Description

Technical Field

[0001] This invention relates to the field of computers, and more specifically to a task matching system and storage medium based on intelligent load balancing. Background Technology

[0002] Matching personnel with tasks is a crucial factor affecting a company's operation and development. In today's big data era, employment services face a massive number of personnel and tasks. How to efficiently match personnel with tasks is an important problem that needs to be solved.

[0003] Currently, matching personnel with tasks mainly relies on computers to perform one-to-one matching, which has low computational efficiency, and there is no effective solution to the load balancing problem of big data cluster computing.

[0004] Therefore, those skilled in the art have provided a task matching system and storage medium based on intelligent load balancing to solve the problems mentioned in the background art. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a task matching system based on intelligent load balancing, comprising the following steps:

[0006] S1. Establish a smart agent matcher matrix;

[0007] S2. Personnel task data initialization;

[0008] S3. Establish the water level eigenvalue matrix;

[0009] S4. Calculation of water level characteristic values;

[0010] S5. The task drift intelligent controller is in operation;

[0011] S6. Results Feedback;

[0012] Preferred configuration: By deploying n servers, each server deploying m intelligent agent matchers, an intelligent agent matcher matrix is ​​established; the servers consist of hardware and software such as CPU, GPU, memory, hard disk, and database; the intelligent agent matcher is a virtual working unit in the server, and its components include CPU, GPU, memory, hard disk, bandwidth, database, and algorithm container.

[0013] Preferred method: By establishing a smart agent matcher matrix The computing resources are defined as follows: i represents the server ID, with a value ranging from (1, n]; j represents the intelligent agent matcher ID, with a value ranging from (1, m]; st represents the status; and us represents the working duration. This is the corresponding matching calculation task number. cp represents the current CPU utilization of this intelligent agent matcher; br represents the current bandwidth utilization of this intelligent agent matcher; me represents the current memory utilization of this intelligent agent matcher.

[0014] Preferred method: By setting a time interval t1, the personnel task list is loaded and initialized every time t1, that is, the personnel task data is initialized.

[0015] Specifically, the personnel and tasks to be matched are loaded into the personnel-task matching table ppl, and matching tasks are randomly loaded from the personnel-task matching table ppl to the intelligent agent matcher matrix.

[0016] The intelligent agent matcher automatically performs matching calculations after completing the loading task.

[0017] The matching calculation uses the cosine similarity algorithm to perform text similarity calculation based on the descriptions of people and tasks, and the calculation results are saved to the database.

[0018] Preferred: Establishing a water level eigenvalue matrix Where i represents the server number, and the value of i ranges from (1, n]; j represents the intelligent agent matcher number, and the value of j ranges from (1, m]; and cv represents the water level feature value.

[0019] Preferred method: Set a time interval t2, and calculate the water level characteristic matrix every time interval t2. The water level characteristic value is calculated to complete the calculation of the water level characteristic value.

[0020] Preferred: The task drift intelligent controller works by setting up k task drift intelligent controllers.

[0021] Preferred: The task drift intelligent controller function performs task drift between intelligent agent matchers to balance the computational tasks among intelligent agent matchers.

[0022] Preferred: A non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the system according to any one of claims 1-5.

[0023] The technical effects and advantages of this invention are as follows:

[0024] This invention discloses a task matching system and storage medium based on intelligent load balancing. By establishing an intelligent agent matcher matrix, personnel tasks are digitized, providing a data foundation for big data computing. By establishing a water level feature value matrix and calculating water level feature values, the working state feature values ​​of the intelligent agent matcher matrix are extracted. Through the relevant calculation methods of the task drift intelligent controller, the efficiency of personnel task matching is improved, intelligent load balancing of big data computing is realized, and the utilization rate of computing resources is improved. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the system structure of this application. Detailed Implementation

[0026] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and design various embodiments with various modifications suitable for a particular purpose.

[0027] Please see Figure 1 This embodiment provides a task matching system based on intelligent load balancing, including the following steps:

[0028] S1. Establish a smart agent matcher matrix;

[0029] S2. Personnel task data initialization;

[0030] S3. Establish the water level eigenvalue matrix;

[0031] S4. Calculation of water level characteristic values;

[0032] S5. The task drift intelligent controller is in operation;

[0033] S6. Results Feedback;

[0034] Preferred configuration: By deploying n servers, each server deploying m intelligent agent matchers, an intelligent agent matcher matrix is ​​established; the servers consist of hardware and software such as CPU, GPU, memory, hard disk, and database; the intelligent agent matcher is a virtual working unit in the server, and its components include CPU, GPU, memory, hard disk, bandwidth, database, and algorithm container.

[0035] By establishing a smart agent matcher matrix The computing resources are defined as follows: i represents the server ID, with a value ranging from (1, n]; j represents the intelligent agent matcher ID, with a value ranging from (1, m]; st represents the status; and us represents the working duration. This is the corresponding matching calculation task number. cp represents the current CPU utilization of this intelligent agent matcher; br represents the current bandwidth utilization of this intelligent agent matcher; me represents the current memory utilization of this intelligent agent matcher.

[0036] By setting a time interval t1, the personnel task list is initialized every time t1, that is, the personnel task data is initialized.

[0037] Specifically, the personnel and tasks to be matched are loaded into the personnel-task matching table ppl, and matching tasks are randomly loaded from the personnel-task matching table ppl to the intelligent agent matcher matrix.

[0038] The intelligent agent matcher automatically performs matching calculations after completing the loading task.

[0039] The matching calculation uses the cosine similarity algorithm to perform text similarity calculation based on the descriptions of people and tasks, and the calculation results are saved to the database.

[0040] Establish water level eigenvalue matrix , where i represents the server number, with a value range of (1, n]; j represents the intelligent agent matcher number, with a value range of (1, m]; and cv represents the water level feature value.

[0041] Set a time interval t2, and calculate the water level characteristic matrix every time interval t2. The water level characteristic value is obtained from the data, and the calculation of the water level characteristic value is completed. The algorithm for calculating the water level characteristic value is as follows:

[0042]

[0043] in, , , Represents the intelligent agent matcher matrix The current CPU utilization, bandwidth utilization, and memory utilization of the intelligent agent matcher. This represents the sum of CPU utilization of all smart agent matchers on the server where the smart agent matcher resides. This represents the sum of bandwidth utilization of all smart agent matchers on the server where the smart agent matcher resides. This represents the sum of memory utilization of all smart agent matchers on the server where the smart agent matcher resides. This indicates the CPU utilization rate of the server. This indicates the bandwidth utilization rate of the server. This indicates the memory utilization rate of the server. This indicates the current CPU utilization of all servers in the system. This indicates the current bandwidth utilization of all servers in the system. This indicates the current memory utilization rate of all servers in the system.

[0044] The task drift intelligent controller works by setting up k task drift intelligent controllers.

[0045] The task drift intelligent controller's function is to perform task drifting between intelligent agent matchers, thereby balancing the computational tasks among them. The task reader of the task drift intelligent controller is a 3x3 matrix, and the task drift intelligent controller uses a water level feature matrix. The system navigates within a given time interval t3, executing a control task every t3 interval. During initial operation, the task drift intelligent controller uses the water level feature matrix... An initial position is randomly generated as the center position, and the water level feature matrix is ​​loaded. Data for a 3x3 area centered at this location was then processed, along with data from nine matrices loaded within the mission drift intelligent controller. The CV values ​​are sorted. A task transfer occurs between the intelligent agent matcher corresponding to the lowest-ranked CV value and the one corresponding to the highest-ranked CV value. The computation task is transferred from the intelligent agent matcher with the highest CV value to the intelligent agent matcher with the lowest CV value. After the task drift intelligent controller completes one cycle, its task reader is cleared, and the controller waits for the next time interval before iteratively executing the computation task.

[0046] The design time interval is t4. Every time interval t4, the matching results between personnel and tasks are displayed on the front end to complete the matching result feedback.

[0047] A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform a system according to any one of claims 1-5.

[0048] Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art and related fields based on the embodiments of the present invention without inventive effort should fall within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described and explained in the present invention, unless otherwise specified or limited, shall be implemented according to conventional means in the art.

Claims

1. A task matching method based on intelligent load balancing, characterized in that, Includes the following steps: S1: Deploy n servers, each server deploys m intelligent agent matchers, establishing an intelligent agent matcher matrix. In this context, i represents the server ID, with a value ranging from (1, n]; j represents the intelligent agent matcher ID, with a value ranging from (1, m]; st represents the status; and us represents the working duration. Here is the corresponding matching calculation task number, cp represents the current CPU utilization of the smart agent matcher, br represents the current bandwidth utilization of the smart agent matcher, and me represents the current memory utilization of the smart agent matcher. S2: Set time interval t1. Every time interval t1, load and initialize the personnel task list to complete the personnel task data initialization task. S3: Establish the water level feature matrix Where i represents the server number, with a value ranging from (1, n]; j represents the intelligent agent matcher number, with a value ranging from (1, m]; and cv represents the water level feature value. The calculation algorithm is as follows: ， in, , , Represents the intelligent agent matcher matrix The current CPU utilization, bandwidth utilization, and memory utilization of the intelligent agent matcher are shown in the figure. This represents the sum of the CPU utilization of all smart agent matchers on the server where the smart agent matcher resides; This represents the sum of bandwidth utilization of all smart agent matchers on the server where the smart agent matcher resides; This represents the sum of memory utilization rates of all smart agent matchers on the server where the smart agent matcher resides; This indicates the CPU utilization of the server; This indicates the bandwidth utilization rate of the server; This indicates the server's memory utilization rate; This indicates the current CPU utilization of all servers in the system. This indicates the current bandwidth utilization of all servers in the system. This indicates the current memory utilization rate of all servers in the system. S4: Set a time interval t2, and calculate the water level characteristic value in the water level characteristic matrix every time interval t2; S5: Set up k task drift intelligent controllers, set the time interval t3, and execute the control task once every time interval t3; S6: Design a time interval t4. Every time interval t4, display the matching results of personnel and tasks on the front end and provide feedback on the completed matching results.

2. The task matching method based on intelligent load balancing according to claim 1, characterized in that, The server includes a CPU, GPU, memory, hard disk, and database.

3. The task matching method based on intelligent load balancing according to claim 1, characterized in that, The intelligent agent matcher is a virtual working unit in the server, and its components include CPU, GPU, memory, hard disk, bandwidth, database, and algorithm container.

4. The task matching method based on intelligent load balancing according to claim 1, characterized in that, During the initialization of the personnel task list loading, the personnel and tasks to be matched are loaded into the personnel task matching table ppl, and matching tasks are randomly loaded from the personnel task matching table ppl to the intelligent agent matcher matrix. The intelligent agent matcher automatically performs matching calculations after completing the loading task; The matching calculation is performed using a cosine similarity algorithm based on the descriptions of the personnel and tasks, and the calculation results are saved to the database.

5. The task matching method based on intelligent load balancing according to claim 1, characterized in that, The task drift intelligent controller's function is to perform task drift between intelligent agent matchers to balance the computational tasks among them; the task reader of the task drift intelligent controller is a 3*3 matrix, and the task drift intelligent controller is located in the water level feature matrix. They wandered around.

6. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method according to any one of claims 1-5.

Images