Data optimization system for economic management

By establishing an intelligent cloud-edge collaborative management and control model in cloud-edge collaborative technology, data security risks and high cost problems caused by uneven quality of edge servers are solved, data processing efficiency and accuracy are improved, and resource allocation is optimized. Reduce operational risks and improve system stability and reliability.

CN118940088AInactive Publication Date: 2024-11-12QILU INST OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410985889.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-23
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In cloud-edge collaboration technology, due to the uneven quality of edge servers during multi-source financial data processing, data security risks are high, and cloud-edge collaboration is relatively high, making it difficult to effectively manage and optimize.

Method used

A data optimization system for economic management was designed. Through data collection, demand assessment, log collection, status assessment and management modules, an intelligent cloud-edge collaborative management model is established to accurately assign core tasks and edge tasks to the most suitable edges. On the server, optimize resource allocation and reduce operational risks.

Benefits of technology

It improves the efficiency and accuracy of data processing, optimizes resource allocation, reduces operational risks, improves the stability and reliability of the system, and reduces manual intervention.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118940088A_ABST
    Figure CN118940088A_ABST
Patent Text Reader

Abstract

The invention discloses a data optimization system for economic management, and relates to the technical field of cloud edge collaboration, and the system comprises a data collection module which collects multi-source financial economic data in real time, classifies the data into core data and edge data, respectively records the processing tasks of the core data and the edge data as a core task and an edge task, and stores the core task and the edge task in a database; the demand evaluation module is used for evaluating the computing power demand of a core task and an edge task, the log collection module collects a maintenance log and an update log of an edge server, the state evaluation module evaluates the risk state and performance change of the server based on the maintenance log and the update log, and the management and control module combines the demand evaluation result and the state evaluation result and controls the operation of the server. And constructing a cloud edge collaborative management and control model, and intelligently allocating tasks to the most suitable edge server. The data processing efficiency and accuracy are improved, resource allocation is optimized, the operation risk is reduced, manual intervention is reduced through automatic management, and the stability and reliability of the whole system are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of cloud-edge collaborative technology, and in particular to a data optimization system for economic management. Background Art

[0002] Multi-source financial data refers to financial-related information from different channels and formats, including stock prices, economic indicators, company financial reports, trading data, news reports, social media content, regulatory documents, user behavior records, sensor data, and third-party research reports. Together, they constitute a comprehensive view of the financial market and provide key data support for financial analysis and decision-making.

[0003] Due to the diversity of multi-source financial data, the amount of data is very large. When dealing with such data, companies usually use cloud-edge collaboration to process and analyze the data. However, the cost of building cloud-edge collaboration is high. In order to save costs, companies usually choose to rent cloud servers and edge servers to process the data. However, the quality of rented edge servers varies, and there are great data security risks during operation. Summary of the invention

[0004] In order to solve the above technical problems, this application proposes the following technical solutions:

[0005] In a first aspect, an embodiment of the present application provides a data optimization system for economic management, including:

[0006] A data collection module, which is used to collect real-time multi-source financial and economic data, classify the multi-source financial and economic data, obtain core data and edge data, record the processing task of the core data as a core task, and record the processing task of the edge data as an edge task;

[0007] A demand assessment module, which is electrically connected to the data acquisition module and is used to assess the computing power requirements of core tasks and edge tasks;

[0008] A log collection module, which is used to obtain the maintenance log and update log of each edge server;

[0009] A status assessment module, the status assessment module is electrically connected to the log collection module, and is used to analyze the abnormal risk status and performance change status of each edge server according to the maintenance log and update log of the edge server;

[0010] The management and control module is electrically connected to the demand assessment module and the status assessment module, and is used to establish a cloud-edge collaborative management and control model according to the abnormal risk status and performance change status of the edge server, and allocate core tasks and edge tasks to different edge servers.

[0011] In a possible implementation, the demand assessment module includes:

[0012] A data conversion unit, the data conversion unit is used to convert the data analog signals in the core task and the edge task into digital signals, and determine the data volume of the core task and the total data volume of the edge task;

[0013] A resource requirement unit, the resource requirement unit is used to analyze the time required for the execution of the basic processing algorithm and the memory space required for the execution according to the basic processing algorithm of the core task and the edge task, and determine the processing resource requirements of the core task and the edge task;

[0014] A resource utilization unit, the resource utilization unit is used to determine the resource utilization of the core task and the edge task according to the maximum number of parallel processing of the core task and the edge task;

[0015] A computing power requirement calculation unit is used to evaluate the computing power requirements of core tasks and edge tasks through a performance requirement formula based on the core task data volume and the overall data volume of edge tasks, the processing resource requirements of the core tasks and edge tasks, and the resource utilization of the core tasks and edge tasks.

[0016] In a possible implementation, the performance requirement formula is:

[0017]

[0018] In the formula, G 总 The total computing power required for core tasks and edge tasks is G 1 For the core task computing power requirements, G 2 is the computing power requirement of the edge task, T is the time required for the execution of the base processing algorithm, M is the memory space required for the execution of the base processing algorithm, P is the maximum number of parallel processing, U is the resource utilization, and k is the label of the core task and the edge task.

[0019] In a possible implementation, the status assessment module includes:

[0020] a risk assessment unit, wherein the risk assessment unit assesses an abnormal risk status of each edge server based on a maintenance log of each edge server;

[0021] A performance prediction unit predicts a performance change state of each edge server based on an update log of each edge server.

[0022] In a possible implementation, the evaluating the abnormal risk status of each edge server based on the maintenance log of each edge server includes:

[0023] Based on the maintenance log of the edge server, abnormal events of the edge server are screened out, wherein the abnormal events include: hardware failure, software error and abnormal access;

[0024] Based on logistic regression, a failure risk probability prediction model of the edge server is established. The failure risk probability prediction model of the edge server is expressed as:

[0025]

[0026] Where h(Y=1|X) is the abnormal risk probability of edge server Y under the condition of historical maintenance log X, X i is the ith abnormal event in the historical maintenance log, e is an exponential function, V represents the total number of abnormal events, α0, α1, α i , α v are the coefficients of the model;

[0027] The failure risk probability prediction model based on edge servers takes the abnormal events of each edge server in the historical maintenance log as input, and takes the abnormal risk probability of the edge server under the influence of the abnormal events of each edge server as output.

[0028] In a possible implementation, predicting the performance change state of each edge server based on the update log of each edge server includes:

[0029] Determine the update time of the update log of the edge server;

[0030] Based on the update log of the edge server, mark the event features that have an impact on the performance, and obtain the iteration event feature data of the edge server, wherein the iteration event features include: hardware iteration and software iteration;

[0031] A performance prediction analysis model is established, and the expression of the performance prediction analysis model is:

[0032] H=β0+β1C+β2E2+…+β j E j +…+β m E m +∈,(j=1,2,...,m)

[0033] Among them, H is the predicted overall performance change state indicator of the edge server, C is the time from the iterative event feature of the edge server to the current time, and E j is the iterative event feature vector of the jth edge server, ∈ is the error term, β0 is the intercept term, β1 is the coefficient of the iteration time, β2, β j …、β mare the eigenvector coefficients of iterative events, and m is the total number of edge servers;

[0034] Based on the performance prediction analysis model, the iterative event feature data of each edge server is used as output, and the quantified performance improvement value of the iterative event feature of the edge server is used as output. According to the time from the iterative event feature of the edge server to the current time, the overall performance change status indicator of the edge server is predicted.

[0035] In a possible implementation, the management and control module includes:

[0036] A grouping unit, wherein the grouping unit classifies edge servers based on abnormal risk status indicator level values ​​of edge servers to obtain a low-risk edge server array and a risk edge server array;

[0037] A task requirement unit, which is used to determine the computing power requirements of core tasks and edge tasks;

[0038] A feature unit, wherein the feature unit is used to use the computing power requirements of the core task and the edge task as feature data, and use the low-risk edge server array and the high-risk edge server array as label data;

[0039] A model building unit, the model building unit is used to build a cloud-edge collaborative management and control model; the cloud-edge collaborative management and control model is expressed as:

[0040]

[0041] Among them, Z is the objective function of the cloud-edge collaborative management model, W r The weight of the tasks to be executed, D xj is a binary decision variable, indicating whether the task r to be executed is assigned to the jth server. is the computing power requirement of the rth task to be executed within the kth (core / edge) task, H j is the changeable state indicator of the jth edge server, L is the total number of tasks to be executed, and m is the total number of edge servers;

[0042] The model prediction unit is based on the cloud-edge collaborative management and control model, takes the computing power requirements of core tasks and edge tasks as feature data as the input of the model, and outputs the label data of low-risk edge servers that meet the computing power requirements of core tasks and high-risk edge servers that meet the computing power requirements of edge tasks.

[0043] In a possible implementation, the model prediction unit includes:

[0044] A first condition subunit, wherein the first condition subunit is based on an overall performance change state indicator of the edge server, and takes satisfying the most core tasks and edge tasks as a first allocation condition;

[0045] The second condition sub-unit uses the smallest abnormal risk state of the core task and the marginal task as the second allocation condition.

[0046] In a possible implementation, the overall performance change state indicator based on the edge server, with the satisfaction of the most core tasks and edge tasks as the first allocation condition, includes:

[0047] Obtain the overall computing power requirements for core tasks and edge tasks;

[0048] Determine the overall performance change status indicators of low-risk edge servers and high-risk edge servers;

[0049] Analyze whether the overall performance change status indicators of all low-risk edge servers meet the overall computing power requirements of core tasks and edge tasks. If so, randomly allocate core tasks and edge tasks; if not, perform a mixed allocation.

[0050] In a possible implementation, taking the smallest abnormal risk state of the core task and the marginal task as the second allocation condition includes:

[0051] Based on the failure risk probability prediction model of edge servers, multiple allocation schemes are substituted into the model to obtain the abnormal risk probability of each allocation scheme;

[0052] According to the abnormal risk probability, the allocation scheme corresponding to the minimum abnormal risk probability is screened out as the final allocation scheme.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] The present invention proposes a data optimization solution for economic management. By establishing an intelligent cloud-edge collaborative management and control model, it can accurately allocate core tasks and edge tasks to the most suitable edge servers, thereby improving the efficiency and accuracy of data processing, optimizing resource allocation, reducing operational risks, and reducing manual intervention through automated management, thereby improving the stability and reliability of the entire system. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 A schematic diagram of the structure of a data optimization system for economic management provided in an embodiment of the present application;

[0056] Figure 2 A schematic diagram of the structure of a demand assessment module provided in an embodiment of the present application;

[0057] Figure 3 A schematic diagram of the structure of a status assessment module provided in an embodiment of the present application;

[0058] Figure 4 A schematic diagram of the structure of the control module provided in the embodiment of the present application. DETAILED DESCRIPTION

[0059] The present solution is described below in conjunction with the accompanying drawings and specific implementation methods.

[0060] See also Figure 1 In this embodiment, a data optimization system for economic management includes: a data acquisition module, a demand assessment module, a log acquisition module, a status assessment module and a management and control module, wherein the data acquisition module is used to collect real-time multi-source financial and economic data, and classify the multi-source financial and economic data into core data and edge data, record the processing tasks of the core data as core tasks, and record the processing tasks of the edge data as edge tasks. The demand assessment module is electrically connected to the data acquisition module, and is used to evaluate the computing power requirements of the core tasks and edge tasks based on the core tasks and edge tasks. The log acquisition module is used to obtain the maintenance log and update log of each edge server. The status assessment module is electrically connected to the log acquisition module, and is used to analyze the abnormal risk status of each edge server and the performance change status of each edge server based on the maintenance log and update log of each edge server. The management and control module is electrically connected to the demand assessment module and the status assessment module, and is used to establish a cloud-edge collaborative management and control model based on the abnormal risk status of the edge server and the performance change status of the edge server, and allocate the core tasks and edge tasks to different edge servers.

[0061] It is understandable that due to the large gap in the status of different edge servers in the cloud-edge collaboration mode, the data is classified into core data (important data) and edge data (unimportant data). When the data is finally distributed, the core data is distributed to the edge server with better status, and the edge data is distributed to the edge server with average status, so as to improve the reliability of cloud-edge collaboration and enhance data security.

[0062] join Figure 2 , a structural diagram of the demand assessment module provided in an embodiment of the present application, the demand assessment module includes: a data conversion unit, a resource demand unit, a resource utilization unit and a computing power demand calculation unit, wherein the data conversion unit is used to convert the data analog signals in the core tasks and the edge tasks into digital signals, and determine the data volume of the core tasks and the total data volume of the edge tasks.

[0063] The resource requirement unit analyzes the time required for the execution of the basic processing algorithms and the memory space required for the execution according to the basic processing algorithms of the core tasks and the edge tasks, and determines the processing resource requirements of the core tasks and the edge tasks.

[0064] The resource utilization unit determines the resource utilization of the core tasks and the edge tasks according to the maximum parallel processing quantity of the core tasks and the edge tasks.

[0065] The computing power requirement calculation unit evaluates the computing power requirements of the core tasks and edge tasks based on the data volume of the core tasks and the overall data volume of the edge tasks, the processing resource requirements of the core tasks and edge tasks, and the resource utilization of the core tasks and edge tasks through the performance requirement formula. The performance requirement formula is as follows:

[0066]

[0067] In the formula, G 总 The total computing power required for core tasks and edge tasks is G 1 For the core task computing power requirements, G 2 is the computing power requirement of the edge task, T is the time required for the execution of the base processing algorithm, M is the memory space required for the execution of the base processing algorithm, P is the maximum number of parallel processing, U is the resource utilization, and k is the sequence number of the core task and the edge task.

[0068] It should be noted that this embodiment provides a basic framework to evaluate and plan system resource requirements, but in actual resource planning, other factors need to be considered, such as task priority, real-time requirements, system redundancy, and fault recovery capability.

[0069] See also Figure 3 , a schematic diagram of the structure of the status assessment module provided in an embodiment of the present application, wherein the status assessment module includes a risk assessment unit and a performance prediction unit. Among them, the risk assessment unit evaluates the abnormal risk status of each edge server based on the maintenance log of each edge server. This embodiment filters out abnormal events of the edge server based on the maintenance log of the edge server, and the abnormal events include hardware failures, software errors, and abnormal access. Based on logistic regression, a failure risk probability prediction model of the edge server is established. The failure risk probability prediction model based on the edge server takes the abnormal events of each edge server in the historical maintenance log as input, and takes the probability of abnormal risk of the edge server under the influence of the abnormal events of each edge server as output.

[0070] Among them, the failure risk probability prediction model expression of the edge server is:

[0071]

[0072] Where h(Y=1|X) is the abnormal risk probability of edge server Y under the condition of historical maintenance log X, and X i is the ith abnormal event in the historical maintenance log, e is an exponential function, V represents the total number of abnormal events, α0, α1, α i , α v are the coefficients of the model.

[0073] The performance prediction unit predicts the performance change state of each edge server based on the update log of each edge server. In this embodiment, the update time of the update log of the edge server is determined, and the associated event features that affect the performance are marked based on the update log of the edge server, and the iteration event feature data of the edge server is obtained. The iteration event features include but are not limited to hardware iteration and software iteration. A performance prediction analysis model is established. Based on the performance prediction analysis model, the iteration event feature data of each edge server is used as input, and the performance improvement value of the iteration event feature of the edge server is quantified as output. According to the time length from the iteration event feature of the edge server to the current time, the overall performance change state indicator of the edge server is predicted.

[0074] Among them, the performance prediction analysis model expression is:

[0075] H=β0+β1C+β2E2+…+β j E j +…+β m E m +∈,(j=1,2,...,m)

[0076] Where H is the predicted overall performance change state indicator of the edge server, C is the time from the iterative event feature of the edge server to the current time, and E j is the feature vector of the jth edge server iteration event, ∈ is the error term, β0 is the intercept term, β1 is the coefficient of the iteration time, β2, β j …、β m are the eigenvector coefficients of iterative events, and m is the total number of edge servers.

[0077] The above embodiment provides reliable decision data for subsequent task allocation by evaluating the abnormal risk index of the edge server and the overall performance change status index of the edge server. Secondly, by analyzing the abnormal risk index of the edge server, the core tasks can be distributed to the edge server with a smaller abnormal risk index in a targeted manner, thereby ensuring the stability and reliability of the overall system.

[0078] See also Figure 4, a schematic diagram of the structure of the control module provided in the embodiment of the present application, the control module includes: a grouping unit, a task requirement unit, a feature unit, a model building unit and a model prediction unit. Among them, the grouping unit classifies the edge servers based on the abnormal risk state indicator level value of the edge servers to obtain a low-risk edge server array and a risk edge server array.

[0079] The task requirement unit is used to determine the computing power requirements of core tasks and edge tasks.

[0080] The feature unit is used to take the computing power requirements of core tasks and edge tasks as feature data, and the low-risk edge server array and the high-risk edge server array as label data.

[0081] The model building unit is used to build a cloud-edge collaborative management and control model. In this embodiment, the cloud-edge collaborative management and control model is expressed as:

[0082]

[0083] Where Z is the objective function of the cloud-edge collaborative management model, W r The weight of the tasks to be executed is uniformly assigned according to the core tasks and edge tasks. xj is a binary decision variable, indicating whether the task r to be executed is assigned to the jth server. is the computing power requirement of the rth task to be executed within the kth (core / edge) task, H j is the changeable state indicator of the jth edge server, L is the total number of tasks to be executed, and m is the total number of edge servers.

[0084] The model prediction unit, based on the cloud-edge collaborative management and control model, takes the computing power requirements of core tasks and edge tasks as feature data as the input of the model, and takes the label data of low-risk edge servers that meet the computing power requirements of core tasks and high-risk edge servers that meet the computing power requirements of edge tasks as output to obtain a task allocation plan.

[0085] This embodiment constructs an objective function with the computing power requirements of core tasks and edge tasks as conditions, maximizes the use of the performance change state indicators of edge servers, and uses the edge server that satisfies the most tasks to be executed as the allocation plan. It ensures that the computing power requirement of each task will not exceed the performance change status indicator of the server to which it is assigned. This ensures that each task can only be assigned to one server.

[0086] Among them, the model prediction unit includes:

[0087] The first condition subunit satisfies the most core tasks and edge tasks as the first allocation condition based on the overall performance change state indicator restriction of the edge server.

[0088] The second condition sub-unit takes the smallest abnormal risk state of the core task and the marginal task as the second allocation condition.

[0089] In this embodiment, based on the overall performance change state indicator restriction of the edge server, the first allocation condition is to satisfy the most core tasks and edge tasks, including:

[0090] Obtain the overall computing power requirements of core tasks and edge tasks, determine the overall performance change status indicators of low-risk edge servers and high-risk edge servers, analyze whether the overall performance change status indicators of all low-risk edge servers meet the overall computing power requirements of core tasks and edge tasks, and if so, randomly allocate the core tasks and edge tasks. If not, perform mixed allocation. If mixed allocation is performed, analyze the performance change status indicators of each low-risk edge server and the performance change status indicators of the high-risk edge server as the upper limit of the running task carrying capacity of the low-risk and high-risk edge servers, traverse and screen the core tasks, define the edge tasks as alternatives, run the core tasks and edge tasks with the upper limit of the running task carrying capacity of the maximum low-risk and high-risk edge servers, and obtain multiple allocation plans.

[0091] The second allocation condition is to use the minimum abnormal risk status of core tasks and marginal tasks, including:

[0092] Based on the failure risk probability prediction model of the edge server, multiple allocation schemes are substituted into the model to obtain the abnormal risk probability of each allocation scheme, and according to the abnormal risk probability, the allocation scheme corresponding to the minimum abnormal risk probability is screened out as the final allocation scheme.

[0093] In this embodiment, a use process of the economic management data optimization system is as follows:

[0094] S101, collect real-time multi-source financial and economic data, and classify the multi-source financial and economic data into core data and edge data, record the processing task of the core data as the core task, and record the processing task of the edge data as the edge task.

[0095] S102, converting the data analog signals in the core task and the edge task into digital signals, and determining the data volume of the core task and the total data volume of the edge task.

[0096] S103, according to the core tasks and edge tasks of the base processing algorithm, analyze the base processing algorithm execution time and execution required memory space, determine the core tasks and edge tasks of the processing resource requirements.

[0097] S104: Determine resource utilization rates of the core tasks and the edge tasks according to the maximum number of parallel processing of the core tasks and the edge tasks.

[0098] S105, based on the data volume of the core tasks and the overall data volume of the edge tasks, the processing resource requirements of the core tasks and the edge tasks, and the resource utilization of the core tasks and the edge tasks, the computing power requirements of the core tasks and the edge tasks are evaluated through the performance requirement formula.

[0099] S106, obtaining the maintenance log and update log of each edge server.

[0100] S107, based on the maintenance log of each edge server, evaluating the abnormal risk status of each edge server.

[0101] S108: predicting the performance change status of each edge server based on the update log of each edge server.

[0102] S109, classifying the edge servers based on the abnormal risk status indicator level values ​​of the edge servers to obtain a low-risk edge server array and a risky edge server array.

[0103] S110, determining the computing power requirements of core tasks and edge tasks.

[0104] S111, taking the computing power requirements of the core tasks and the edge tasks as feature data, and taking the low-risk edge server array and the high-risk edge server array as label data.

[0105] S112, build a cloud-edge collaborative management and control model.

[0106] S113, based on the cloud-edge collaborative management and control model, takes the computing power requirements of core tasks and edge tasks as feature data as the model input, and takes the label data of low-risk edge servers that meet the computing power requirements of core tasks and high-risk edge servers that meet the computing power requirements of edge tasks as output to obtain a task allocation plan.

[0107] To sum up, this embodiment, by establishing an intelligent cloud-edge collaborative management and control model, can accurately allocate core tasks and edge tasks to the most suitable edge servers, thereby improving the efficiency and accuracy of data processing, optimizing resource allocation, reducing operational risks, and reducing manual intervention through automated management, thereby improving the stability and reliability of the entire system.

[0108] In the embodiments of the present application, "at least one" means one or more, and "more" means two or more. "And / or" describes the association relationship of the associated objects, indicating that three relationships may exist. For example, A and / or B can mean that A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship.

[0109] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0110] The above is only a specific implementation of the present application. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. The protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A data optimization system for economic management, characterized in that: include: A data collection module, which is used to collect real-time multi-source financial and economic data, classify the multi-source financial and economic data, obtain core data and edge data, record the processing task of the core data as a core task, and record the processing task of the edge data as an edge task; A demand assessment module, which is electrically connected to the data acquisition module and is used to assess the computing power requirements of core tasks and edge tasks; A log collection module, which is used to obtain the maintenance log and update log of each edge server; A status assessment module, the status assessment module is electrically connected to the log collection module, and is used to analyze the abnormal risk status and performance change status of each edge server according to the maintenance log and update log of the edge server; The management and control module is electrically connected to the demand assessment module and the status assessment module, and is used to establish a cloud-edge collaborative management and control model according to the abnormal risk status and performance change status of the edge server, and allocate core tasks and edge tasks to different edge servers.

2. The economic management data optimization system according to claim 1, characterized in that: The needs assessment module includes: A data conversion unit, the data conversion unit is used to convert the data analog signals in the core task and the edge task into digital signals, and determine the data volume of the core task and the total data volume of the edge task; A resource requirement unit, the resource requirement unit is used to analyze the time required for the execution of the basic processing algorithm and the memory space required for the execution according to the basic processing algorithm of the core task and the edge task, and determine the processing resource requirements of the core task and the edge task; A resource utilization unit, the resource utilization unit is used to determine the resource utilization of the core task and the edge task according to the maximum number of parallel processing of the core task and the edge task; A computing power requirement calculation unit is used to evaluate the computing power requirements of core tasks and edge tasks through a performance requirement formula based on the core task data volume and the overall data volume of edge tasks, the processing resource requirements of the core tasks and edge tasks, and the resource utilization of the core tasks and edge tasks.

3. The economic management data optimization system according to claim 2, characterized in that: The performance requirement formula is: In the formula, G 总 The total computing power required for core tasks and edge tasks is G 1 For the core task computing power requirements, G 2 is the computing power requirement of the edge task, T is the time required for the execution of the base processing algorithm, M is the memory space required for the execution of the base processing algorithm, P is the maximum number of parallel processing, U is the resource utilization, and k is the label of the core task and the edge task.

4. The economic management data optimization system according to claim 1, characterized in that: The status assessment module includes: a risk assessment unit, wherein the risk assessment unit assesses an abnormal risk status of each edge server based on a maintenance log of each edge server; A performance prediction unit predicts a performance change state of each edge server based on an update log of each edge server.

5. The economic management data optimization system according to claim 4, characterized in that: The step of evaluating the abnormal risk status of each edge server based on the maintenance log of each edge server includes: Based on the maintenance log of the edge server, abnormal events of the edge server are screened out, wherein the abnormal events include: hardware failure, software error and abnormal access; Based on logistic regression, a failure risk probability prediction model of the edge server is established. The failure risk probability prediction model of the edge server is expressed as: Where h(Y=1|X) is the abnormal risk probability of edge server Y under the condition of historical maintenance log X, X i is the ith abnormal event in the historical maintenance log, e is an exponential function, V represents the total number of abnormal events, α0, α1, α i , α v are the coefficients of the model; The failure risk probability prediction model based on edge servers takes the abnormal events of each edge server in the historical maintenance log as input, and takes the abnormal risk probability of the edge server under the influence of the abnormal events of each edge server as output.

6. The economic management data optimization system according to claim 4, characterized in that: The predicting of the performance change status of each edge server based on the update log of each edge server includes: Determine the update time of the update log of the edge server; Based on the update log of the edge server, mark the event features that have an impact on the performance, and obtain the iteration event feature data of the edge server, wherein the iteration event features include: hardware iteration and software iteration; A performance prediction analysis model is established, and the expression of the performance prediction analysis model is: H=β0+β1C+β2E2+…+β j E j +…+b m E m +∈,(j=1,2,...,m) Among them, H is the predicted overall performance change status indicator of the edge server, C is the time from the iterative event feature of the edge server to the current time, and E j is the iterative event feature vector of the jth edge server, ∈ is the error term, β0 is the intercept term, β1 is the coefficient of the iteration time, β2, β j …、β m are the eigenvector coefficients of iterative events, and m is the total number of edge servers; Based on the performance prediction analysis model, the iterative event feature data of each edge server is used as output, and the quantified performance improvement value of the iterative event feature of the edge server is used as output. According to the time from the iterative event feature of the edge server to the current time, the overall performance change status indicator of the edge server is predicted.

7. The economic management data optimization system according to claim 1, characterized in that: The control module includes: A grouping unit, wherein the grouping unit classifies edge servers based on abnormal risk status indicator level values ​​of edge servers to obtain a low-risk edge server array and a risk edge server array; A task requirement unit, which is used to determine the computing power requirements of core tasks and edge tasks; A feature unit, wherein the feature unit is used to use the computing power requirements of the core task and the edge task as feature data, and use the low-risk edge server array and the high-risk edge server array as label data; A model building unit, the model building unit is used to build a cloud-edge collaborative management and control model; the cloud-edge collaborative management and control model is expressed as: Among them, Z is the objective function of the cloud-edge collaborative management model, W r The weight of the tasks to be executed, D xj is a binary decision variable, indicating whether the task r to be executed is assigned to the jth server. r k is the computing power requirement of the rth task to be executed within the kth (core / edge) task, H j is the changeable state indicator of the jth edge server, L is the total number of tasks to be executed, and m is the total number of edge servers; The model prediction unit is based on the cloud-edge collaborative management and control model, takes the computing power requirements of core tasks and edge tasks as feature data as the input of the model, and outputs the label data of low-risk edge servers that meet the computing power requirements of core tasks and high-risk edge servers that meet the computing power requirements of edge tasks.

8. The economic management data optimization system according to claim 7, characterized in that: The model prediction unit comprises: A first condition subunit, wherein the first condition subunit is based on an overall performance change state indicator of the edge server, and takes satisfying the most core tasks and edge tasks as a first allocation condition; The second condition sub-unit uses the smallest abnormal risk state of the core task and the marginal task as the second allocation condition.

9. The economic management data optimization system according to claim 8, characterized in that: The overall performance change state indicator based on the edge server, with the largest number of core tasks and edge tasks as the first allocation condition, includes: Obtain the overall computing power requirements for core tasks and edge tasks; Determine the overall performance change status indicators of low-risk edge servers and high-risk edge servers; Analyze whether the overall performance change status indicators of all low-risk edge servers meet the overall computing power requirements of core tasks and edge tasks. If so, randomly allocate core tasks and edge tasks; if not, perform a mixed allocation.

10. The economic management data optimization system according to claim 8, characterized in that: The second allocation condition of taking the smallest abnormal risk state of the core task and the marginal task as the second allocation condition includes: Based on the failure risk probability prediction model of edge servers, multiple allocation schemes are substituted into the model to obtain the abnormal risk probability of each allocation scheme; According to the abnormal risk probability, the allocation scheme corresponding to the minimum abnormal risk probability is screened out as the final allocation scheme.