Resource Information Performance Evaluation Management System and Methods for Big Data

Abstract

This invention belongs to the field of resource efficiency evaluation technology. It discloses a resource information efficiency evaluation management system and method for big data, including a resource panorama modeling module. This module performs factor-based modeling of the resources to be evaluated, decomposing resource information into functional factors and constraint factors to form a resource panorama efficiency vector. Based on this vector, it calculates the efficiency of individual resources. A resource interaction and association module constructs a resource interaction and association network based on the efficiency results of individual resources, classifying resource edges into gain edges and hindrance edges. By propagating the efficiency of individual resources within this network, it obtains the overall system efficiency. An efficiency balance baseline module dynamically monitors the overall system efficiency, statistically analyzing the overall efficiency mean and fluctuation range using a sliding time window to establish an adaptive efficiency baseline. The baseline deviation magnitude is used to determine whether the current system efficiency status needs optimization. This improves the scientific rigor and dynamism of resource information efficiency evaluation.

Description

Technical Field

[0001] This invention relates to the field of resource efficiency assessment technology, and more specifically, to a resource information efficiency assessment management system and method for big data. Background Technology

[0002] Existing resource information performance evaluation management systems and methods for big data mainly suffer from the following problems:

[0003] With the rapid development of big data technology, the types and scale of resources are constantly increasing. How to evaluate and rationally allocate the performance of multi-source heterogeneous resources has become a crucial issue in resource management. Existing resource information performance evaluation management systems and methods for big data typically suffer from the following shortcomings in the evaluation and allocation process:

[0004] Traditional methods often focus only on the performance indicators or correlation analysis of single resources, making it difficult to directly quantify the positive or negative impact of multiple resources acting simultaneously on the overall system performance. They lack clear modeling methods for complex gain and hindrance relationships between resources, often relying on correlation coefficients or empirical rules to establish resource relationships. This approach fails to effectively distinguish between positive gain and negative hindrance effects and lacks a weighted quantification mechanism based on performance contribution, thus making it difficult to accurately characterize the interaction effects between resources. Furthermore, existing methods cannot form a visualized resource interaction network at the system level, making it difficult to intuitively represent complex dependencies and affecting the accuracy of resource optimization decisions.

[0005] The sheer number and diverse combinations of resources mean that traditional experimental or simulation methods often fail to cover all resource combinations, leading to incomplete estimations of resource synergy effects. Furthermore, these methods cannot achieve rapid and scalable resource synergy effect assessments in big data environments, limiting the reliability and real-time nature of system performance evaluation results. Existing resource scheduling methods typically allocate resources based solely on performance indicators or usage priorities, making it difficult to guarantee optimal overall system-level performance.

[0006] When faced with multiple available resources, allocation is often done piecemeal or by manual experience, lacking the ability to systematically optimize resource combinations under budget constraints. This can easily lead to the underutilization of some high-value resources, resulting in wasted resources. Existing systems also suffer from deficiencies in the transparency and interpretability of scheduling decisions, making it difficult to clearly demonstrate the rationality of resource selection to users, thus affecting the credibility of resource management results and their widespread application. Therefore, this invention proposes a resource information efficiency evaluation management system and method oriented towards big data to address the aforementioned problems. Summary of the Invention

[0007] To overcome the aforementioned shortcomings of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a resource information efficiency evaluation and management system for big data, comprising:

[0008] The resource panorama modeling module performs factor-based modeling of the resources to be evaluated, decomposing resource information into functional factors and constraint factors to form a resource panorama effectiveness vector; based on the resource panorama effectiveness vector, the effectiveness of a single resource is calculated to obtain the single resource effectiveness result;

[0009] The resource interaction and association module constructs a resource interaction and association network between resources based on the single resource performance results. It divides the edges between resources into gain edges and hindrance edges. By propagating the single resource performance in the resource interaction and association network, the overall system performance is obtained.

[0010] The performance balance baseline module dynamically monitors the overall system performance, statistically analyzes the overall performance mean and fluctuation range based on a sliding time window, and establishes an adaptive performance baseline; it judges whether the current system performance status needs optimization based on the baseline deviation magnitude;

[0011] The efficiency-first scheduling module, if optimization is required, collects the marginal efficiency contribution and cost consumption of resources, calculates the efficiency return rate, and prioritizes the resources; based on the priority ranking results, it executes resource scheduling under budget constraints.

[0012] Specifically, the method for forming the resource panorama performance vector includes:

[0013] Information collection and standardization processing are carried out on the resources to be evaluated. The collected resource information includes resource attribute information, usage status information, historical performance information and environmental condition information. The units of the collected resource information are standardized, missing resource information is filled by interpolation, and outliers in the resource information are removed by standard deviation detection method.

[0014] The standardized resource information is decomposed into functional factors and constraint factors. The functional factors represent the positive performance contribution of the resource in completing the task, and the constraint factors represent the negative consumption generated during the use of the resource. The functional factors and constraint factors are combined into a vector in a predefined order to form a resource panoramic performance vector. Each dimension of the panoramic performance vector corresponds to a different functional factor and constraint factor.

[0015] Specifically, the method for obtaining the single resource performance result includes:

[0016] The constructed resource panorama efficiency vector is used as input, and the functional factors and constraint factors in the resource panorama efficiency vector are normalized. The positive efficiency contribution of the resource is calculated based on the functional factors. The functional factors are quantified by the weighted average method, and the comprehensive positive efficiency value of each functional factor is obtained according to the preset functional factor weight.

[0017] The negative consumption of resources is calculated based on constraint factors. The constraint factors are quantified through a cost function, and each constraint factor obtains a comprehensive negative consumption value according to the preset constraint factor weight. The positive efficiency contribution and negative consumption are combined to obtain the single resource efficiency value. The calculated single resource efficiency value is recorded as the single resource efficiency result.

[0018] Specifically, the method for constructing a resource interaction and association network among resources includes:

[0019] All resource records with calculated single-resource effectiveness results in the system are organized into a resource set. Any resource pair is selected from the resource set, and the joint effectiveness when the two resources act simultaneously is calculated based on the Monte Carlo simulation method. The resource synergy effectiveness difference is calculated based on the joint effectiveness and single-resource effectiveness results, and the positive and negative effects between the two resources are quantified through the resource synergy effectiveness difference.

[0020] Based on the difference in resource synergy effectiveness, resource edges are classified. When the difference in resource synergy effectiveness is greater than zero, resource pairs are defined as gain edges; when the difference in resource synergy effectiveness is less than zero, resource pairs are defined as hindrance edges; when the difference in resource synergy effectiveness is equal to zero, it indicates that there is no synergy effect between resources, and no edge is established. All resources are treated as network nodes, and gain edges and hindrance edges are used as weighted edges to connect the nodes, forming a positive and negative differentiated resource interaction and association network.

[0021] Specifically, the method for obtaining the overall system performance includes:

[0022] The single-resource performance result corresponding to each node in the resource interaction association network is used as the initial performance of the node; the performance is propagated for each edge in the resource interaction association network, and the node performance is updated according to the edge weight;

[0023] The efficiency propagation step is repeated, iterating over all nodes in the resource interaction network until a preset number of iterations is reached; the overall system efficiency is calculated by weighted summation of the efficiency of all nodes in the resource interaction network.

[0024] Specifically, the method for establishing an adaptive performance baseline includes:

[0025] The overall system performance is recorded in chronological order to form a system performance time series; the length of the sliding time window is preset, and the overall system performance within the most recent sliding time window is used as the analysis object; within the sliding time window, the mean and fluctuation range of the overall system performance are calculated.

[0026] Based on the mean and fluctuation range of the overall system performance, the mean of the overall system performance within the sliding time window is used as the center value of the adaptive performance baseline, and the fluctuation range is used as the dynamic fluctuation interval to construct the adaptive performance baseline interval.

[0027] Specifically, the method for determining whether the current performance state of the system needs optimization based on the baseline deviation magnitude includes:

[0028] A preset baseline deviation threshold is set, and the baseline deviation is calculated based on the average of the overall system performance at the current time point and the overall system performance within the sliding time window. When the baseline deviation exceeds the preset baseline deviation threshold, the system is determined to be in an unsafe or abnormal state and needs to be optimized.

[0029] Specifically, the method for prioritizing resources includes:

[0030] If it is determined that the current performance status of the system needs to be optimized, the marginal performance contribution and cost consumption of each resource in the system are collected; the performance return rate of each resource is obtained by calculating the ratio of the marginal performance contribution of each resource to the corresponding cost consumption; and all resources in the system are sorted in descending order according to the performance return rate to form a resource priority list.

[0031] Specifically, the method for performing resource scheduling under budget constraints includes:

[0032] Define the total efficiency return of the selected resource set and define the total cost variable. Under the condition that the total cost variable does not exceed the budget constraint, select the optimal resource set to maximize the total efficiency return. Based on the optimal selection result, schedule and allocate the resources.

[0033] Resource information effectiveness evaluation and management methods for big data include:

[0034] S1. Perform factorization modeling on the resource to be evaluated, decompose the resource information into functional factors and constraint factors to form a resource panoramic efficiency vector; calculate the efficiency of a single resource based on the resource panoramic efficiency vector to obtain the single resource efficiency result.

[0035] S2. Construct a resource interaction and association network between resources based on the single resource performance results, divide the resource edges into gain edges and hindrance edges, and obtain the overall system performance by propagating the single resource performance in the resource interaction and association network.

[0036] S3. Dynamically monitor the overall system performance, statistically analyze the overall performance mean and fluctuation range based on the sliding time window, and establish an adaptive performance baseline; determine whether the current system performance needs optimization based on the baseline deviation.

[0037] S4. If optimization is required, collect the marginal performance contribution and cost consumption of resources, calculate the performance return rate, and prioritize the resources; based on the priority ranking results, execute resource scheduling under budget constraints.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] This invention achieves a quantitative description of the positive and negative effects between resources by calculating the joint effectiveness of resource pairs based on Monte Carlo simulation and combining the results of single resource effectiveness to calculate the synergistic effectiveness difference. Compared with traditional methods, it can not only distinguish between gain and hindrance effects but also obtain clear numerical quantification indicators. Based on the synergistic effectiveness difference, it divides resources into gain edges, hindrance edges, or no edges, constructing a network with resource nodes and weighted edges to achieve a structured representation of complex dependencies between resources. It can intuitively present the differences in positive and negative effects between resources and can be analyzed and optimized at the network level. It can handle hundreds or thousands of resources and their combinations in big data systems, meeting the resource effectiveness management needs of big data applications. Through the quantified synergistic effectiveness difference and the positive and negative differentiation network structure, it can identify efficient resource combinations and potential hindrance resources, providing a scientific basis for system effectiveness optimization, priority scheduling, and resource reconfiguration. This improves the accuracy and efficiency of overall system effectiveness monitoring, analysis, and optimization.

[0040] By collecting resource efficiency and cost information and establishing a quantitative relationship between the two, the scheduling process can achieve a dynamic balance between efficiency improvement and budget consumption. By calculating the total efficiency and cost of the resource set, the system automatically selects the optimal resource combination under budget constraints, ensuring that limited budget investment yields maximum system efficiency improvement. Under multiple resource combinations, the relationship between efficiency and cost can be directly compared, proving that the final selected combination is indeed optimal, thereby improving the transparency and reliability of system scheduling. The system can automatically calculate, combine, and prioritize resource efficiency, reducing manual intervention and improving the efficiency and intelligence of resource management. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the resource information efficiency evaluation and management system for big data according to the present invention.

[0042] Figure 2 This is a schematic diagram of the resource information efficiency evaluation and management method for big data according to the present invention;

[0043] Figure 3 This is a schematic diagram of the method for constructing a resource interaction and association network between resources provided by the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example

[0045] Please see Figure 1 and Figure 3 As shown, this embodiment provides a resource information efficiency evaluation and management system for big data, specifically including the following steps:

[0046] The resource panorama modeling module performs factor-based modeling of the resources to be evaluated, decomposing resource information into functional factors and constraint factors to form a resource panorama effectiveness vector; based on the resource panorama effectiveness vector, the effectiveness of a single resource is calculated to obtain the single resource effectiveness result;

[0047] The resource interaction and association module constructs a resource interaction and association network between resources based on the single resource performance results. It divides the edges between resources into gain edges and hindrance edges. By propagating the single resource performance in the resource interaction and association network, the overall system performance is obtained.

[0048] The performance balance baseline module dynamically monitors the overall system performance, statistically analyzes the overall performance mean and fluctuation range based on a sliding time window, and establishes an adaptive performance baseline; it judges whether the current system performance status needs optimization based on the baseline deviation magnitude;

[0049] The efficiency-first scheduling module, if optimization is required, collects the marginal efficiency contribution and cost consumption of resources, calculates the efficiency return rate, and prioritizes the resources; based on the priority ranking results, it executes resource scheduling under budget constraints.

[0050] Methods for generating resource panorama performance vectors include:

[0051] Information collection and standardization processing are carried out on the resources to be evaluated. The collected resource information includes resource attribute information, usage status information, historical performance information and environmental condition information. The units of the collected resource information are standardized, missing resource information is filled by interpolation, and outliers in the resource information are removed by standard deviation detection method.

[0052] The standardized resource information is decomposed into functional factors and constraint factors. The functional factors represent the positive performance contribution of the resource in completing the task, and the constraint factors represent the negative consumption generated during the use of the resource. The functional factors and constraint factors are combined into a vector in a predefined order to form a resource panoramic performance vector. Each dimension of the panoramic performance vector corresponds to a different functional factor and constraint factor.

[0053] Methods for obtaining single-resource performance results include:

[0054] The constructed resource panorama efficiency vector is used as input, and the functional factors and constraint factors in the resource panorama efficiency vector are normalized. The positive efficiency contribution of the resource is calculated based on the functional factors. The functional factors are quantified by the weighted average method, and the comprehensive positive efficiency value of each functional factor is obtained according to the preset functional factor weight.

[0055] For example, the positive efficiency contribution of resources is: ;in, Indicates the positive contribution of resources to efficiency; Indicates the first The value of each constraint factor, that is, the positive performance index corresponding to the functional factor, such as processing capacity, speed, and efficiency; Indicates the preset number The weights of each functional factor; Indicates the total number of functional factors; Index representing the functional factor;

[0056] The negative consumption of resources is calculated based on constraint factors. The constraint factors are quantified through a cost function, and each constraint factor obtains a comprehensive negative consumption value according to the preset constraint factor weight.

[0057] For example, the overall negative consumption value can be expressed as: ;in, This indicates the overall negative consumption value; Indicates the first The value of a constraint factor, that is, the negative consumption value corresponding to the constraint factor, such as the cost of use; Indicates the preset number The weights of each constraint factor; Indicates the total number of constraint factors; The index represents the constraint factor; the positive performance contribution and negative consumption are combined for calculation to obtain the single resource performance value; the single resource performance value is: ;in, This represents the single resource efficiency value; the calculated single resource efficiency value is recorded as the single resource efficiency result.

[0058] Methods for constructing resource interaction networks among resources include:

[0059] All resource records with calculated single-resource effectiveness results in the system are organized into a resource set. Any resource pair is selected from the resource set, and the joint effectiveness when the two resources act simultaneously is calculated based on the Monte Carlo simulation method. The resource synergy effectiveness difference is calculated based on the joint effectiveness and single-resource effectiveness results, and the positive and negative effects between the two resources are quantified through the resource synergy effectiveness difference.

[0060] Poor resource synergy efficiency is: ;in, Representing resources and Poor resource synergy efficiency between them; Representing resources and The combined effectiveness when acting simultaneously; Representing resources The single-resource efficiency results; Representing resources The single-resource efficiency results; Represents the first in the resource set One resource object; Represents the first in the resource set One resource object;

[0061] Based on the difference in resource synergy effectiveness, resource edges are classified. When the difference in resource synergy effectiveness is greater than zero, resource pairs are defined as gain edges; when the difference in resource synergy effectiveness is less than zero, resource pairs are defined as hindrance edges; when the difference in resource synergy effectiveness is equal to zero, it indicates that there is no synergy effect between resources, and no edge is established. All resources are treated as network nodes, and gain edges and hindrance edges are used as weighted edges to connect the nodes, forming a positive and negative differentiated resource interaction and association network.

[0062] Methods for obtaining overall system performance include:

[0063] The single-resource performance result of each node in the resource interaction association network is used as the initial performance of the node; performance propagation is performed on each edge in the resource interaction association network, where the gain edge amplifies the performance of the source node to the target node, and the blocking edge inhibits the performance of the source node to the target node. The node performance is updated according to the edge weight.

[0064] The efficiency propagation step is repeated, iterating over all nodes in the resource interaction network until a preset number of iterations is reached; the overall system efficiency is calculated by weighted summation of the efficiency of all nodes in the resource interaction network.

[0065] Methods for establishing adaptive performance baselines include:

[0066] The overall system performance is recorded in chronological order to form a system performance time series; the length of the sliding time window is preset, and the overall system performance within the most recent sliding time window is used as the analysis object; within the sliding time window, the mean and fluctuation range of the overall system performance are calculated.

[0067] The mean of the overall system performance is obtained by summing the overall system performance within the sliding time window and dividing by the number of overall system performances. The mean is used to characterize the current performance level. The fluctuation range is obtained by calculating the standard deviation of the overall system performance, which characterizes the degree of performance fluctuation.

[0068] Based on the mean and fluctuation range of the overall system performance, the mean of the overall system performance within the sliding time window is used as the center value of the adaptive performance baseline, and the fluctuation range is used as the dynamic fluctuation range to construct the adaptive performance baseline range. The adaptive performance baseline range is dynamically adjusted as the sliding time window is updated.

[0069] The adaptive performance baseline range is: ;in, This represents the average overall system performance within the sliding time window; This represents the sensitivity factor, used to adjust the baseline bandwidth for adaptive performance; This indicates the range of fluctuation in the overall system performance within the sliding time window.

[0070] Methods for determining whether the current system performance needs optimization based on baseline deviation include:

[0071] A preset baseline deviation threshold is set, and the baseline deviation is calculated based on the average of the overall system performance at the current time point and the overall system performance within the sliding time window. When the baseline deviation exceeds the preset baseline deviation threshold, the system is determined to be in an unsafe or abnormal state and needs to be optimized.

[0072] The baseline deviation is: ;in, Indicates the current time point Overall system efficiency; Indicates the index of the current time point;

[0073] The preset baseline deviation threshold is: ;in, This indicates the preset baseline deviation threshold. This represents the warning trigger coefficient, used to adjust the system's sensitivity to deviations. The larger the coefficient, the higher the system's tolerance for performance deviations; the smaller the coefficient, the easier it is for the system to trigger a warning.

[0074] Methods for prioritizing resources include:

[0075] If it is determined that the current performance status of the system needs to be optimized, the marginal performance contribution and cost consumption of each resource in the system are collected; the performance return rate of each resource is obtained by calculating the ratio of the marginal performance contribution of each resource to the corresponding cost consumption; all resources in the system are sorted in descending order according to the performance return rate to form a resource priority list, and resources with higher performance return rates have higher priority in the sorting.

[0076] It should be noted that the marginal utility contribution of each resource in the system is calculated by randomly simulating resource combinations in the resource interaction network and statistically analyzing the expected incremental contribution of each resource to the overall system utility. The cost consumption of each resource in the system is obtained as follows: for each resource, cost consumption information during use is collected, this information is standardized, and a weighted average is calculated based on preset weights for each consumption indicator in the cost consumption information to obtain the cost consumption of each resource in the system; the cost consumption information includes usage costs, energy consumption, and maintenance costs.

[0077] The return on efficiency for each resource is: ;in, Indicates the first The return on efficiency of a resource; Indicates the first The marginal utility contribution of each resource; Indicates the first The cost of each resource; An index representing a resource;

[0078] Methods for performing resource scheduling under budget constraints include:

[0079] Define the total efficiency return of the selected resource set and define the total cost variable. Under the condition that the total cost variable does not exceed the budget constraint, select the optimal resource set to maximize the total efficiency return. Based on the optimal selection result, schedule and allocate the resources.

[0080] The overall performance return of the selected resource set is: ;in, Representing resources Whether to be selected for scheduling A value of 1 indicates that resources are being scheduled. , A value of 0 indicates that no resources are scheduled. The total cost variable is: ;

[0081] For example, suppose the system has 3 resources:

[0082] Resource 1: Performance Return The cost is 5;

[0083] Resource 2: Performance Return The cost is 4, and the value is 6.

[0084] Resource 3: Performance Return The value is 10, and the cost is 7.

[0085] The budget constraint is 10; if resource 1 is selected, then... =1, otherwise The value is 0; if resource 2 is selected, then... =1, otherwise The value is 0; if resource 3 is selected, then... =1, otherwise =0;

[0086] The overall performance return of the selected resource set is: ;

[0087] The total cost variable is: ;

[0088] If resources 1 and 2 are selected: then the overall efficiency return of the selected resource set is The total cost variable is: ,feasible;

[0089] If resources 1 and 3 are selected: then the overall performance return of the selected resource set is The total cost variable is: Exceeding the budget;

[0090] If resources 2 and 3 are selected: then the overall efficiency return of the selected resource set is The total cost variable is: Exceeding the budget;

[0091] If only resource 3 is selected: then the overall efficiency return of the selected resource set is The total cost variable is: ,feasible;

[0092] Optimal solution: Select resources 1 and 2. The total efficiency return of the selected resource set is 14, and the budget constraint is satisfied.

[0093] This embodiment calculates the joint effectiveness of resource pairs using Monte Carlo simulation and combines the results of single resource effectiveness to calculate the synergistic effectiveness difference, thus achieving a quantitative description of the positive and negative effects between resources. Compared with traditional methods, it can not only distinguish between gain and hindrance effects but also obtain clear numerical quantification indicators. Based on the synergistic effectiveness difference, it divides resources into gain edges, hindrance edges, or no edges, constructing a network with resource nodes and weighted edges to achieve a structured representation of complex dependencies between resources. It can intuitively present the differences in positive and negative effects between resources and can be analyzed and optimized at the network level. It can handle hundreds or thousands of resources and their combinations in big data systems, meeting the resource effectiveness management needs of big data applications. Through the quantified synergistic effectiveness difference and the positive-negative differentiation network structure, it can identify efficient resource combinations and potential hindrance resources, providing a scientific basis for system effectiveness optimization, priority scheduling, and resource reconfiguration. This improves the accuracy and efficiency of overall system effectiveness monitoring, analysis, and optimization.

[0094] By collecting resource efficiency and cost information and establishing a quantitative relationship between the two, the scheduling process can achieve a dynamic balance between efficiency improvement and budget consumption. By calculating the total efficiency and cost of the resource set, the system automatically selects the optimal resource combination under budget constraints, ensuring that limited budget investment yields maximum system efficiency improvement. Under multiple resource combinations, the relationship between efficiency and cost can be directly compared, proving that the final selected combination is indeed optimal, thereby improving the transparency and reliability of system scheduling. The system can automatically calculate, combine, and prioritize resource efficiency, reducing manual intervention and improving the efficiency and intelligence of resource management. Example

[0095] Please see Figure 2 As shown, the parts not described in detail in this embodiment are described in Embodiment 1. A resource information efficiency evaluation and management method for big data is provided, including:

[0096] S1. Perform factorization modeling on the resource to be evaluated, decompose the resource information into functional factors and constraint factors to form a resource panoramic efficiency vector; calculate the efficiency of a single resource based on the resource panoramic efficiency vector to obtain the single resource efficiency result.

[0097] S2. Construct a resource interaction and association network between resources based on the single resource performance results, divide the resource edges into gain edges and hindrance edges, and obtain the overall system performance by propagating the single resource performance in the resource interaction and association network.

[0098] S3. Dynamically monitor the overall system performance, statistically analyze the overall performance mean and fluctuation range based on the sliding time window, and establish an adaptive performance baseline; determine whether the current system performance needs optimization based on the baseline deviation.

[0099] S4. If optimization is required, collect the marginal performance contribution and cost consumption of resources, calculate the performance return rate, and prioritize the resources; based on the priority ranking results, execute resource scheduling under budget constraints.

[0100] Since the electronic device described in this embodiment is the electronic device used to implement the resource information efficiency evaluation management system and method for big data in the embodiments of this application, those skilled in the art can understand the specific implementation method and various variations of the electronic device in this embodiment based on the resource information efficiency evaluation management system and method for big data described in the embodiments of this application. Therefore, how the electronic device implements the method in the embodiments of this application will not be described in detail here. Any electronic device used by those skilled in the art to implement the resource information efficiency evaluation management system and method for big data in the embodiments of this application falls within the scope of protection of this application.

[0101] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0102] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for users of ordinary technical skills, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A resource information efficiency evaluation and management system for big data, characterized in that, include: The resource panorama modeling module performs factor-based modeling of the resources to be evaluated, decomposing resource information into functional factors and constraint factors to form a resource panorama effectiveness vector. Based on the resource panoramic efficiency vector, the efficiency of a single resource is calculated to obtain the single resource efficiency result; The method for forming the resource panorama efficiency vector includes: Information collection and standardization processing are carried out on the resources to be evaluated. The collected resource information includes resource attribute information, usage status information, historical performance information and environmental condition information. The units of the collected resource information are standardized, missing resource information is filled by interpolation, and outliers in the resource information are removed by standard deviation detection method. The standardized resource information is decomposed into functional factors and constraint factors. The functional factors represent the positive performance contribution of the resource in completing the task, and the constraint factors represent the negative consumption generated during the use of the resource. The functional factors and constraint factors are combined into a vector in a predefined order to form a resource panoramic performance vector. Each dimension of the panoramic performance vector corresponds to a different functional factor and constraint factor. The resource interaction and association module constructs a resource interaction and association network between resources based on the single resource performance results. It divides the edges between resources into gain edges and hindrance edges. By propagating the single resource performance in the resource interaction and association network, the overall system performance is obtained. The method for constructing a resource interaction and association network among resources includes: All resource records with calculated single-resource effectiveness results in the system are organized into a resource set. Any resource pair is selected from the resource set, and the joint effectiveness when the two resources act simultaneously is calculated based on the Monte Carlo simulation method. The resource synergy effectiveness difference is calculated based on the joint effectiveness and single-resource effectiveness results, and the positive and negative effects between the two resources are quantified through the resource synergy effectiveness difference. Resource edges are categorized based on the difference in resource synergy effectiveness. When the difference in resource synergy effectiveness is greater than zero, resource pairs are defined as gain edges; when the difference in resource synergy effectiveness is less than zero, resource pairs are defined as hindrance edges; when the difference in resource synergy effectiveness is equal to zero, it indicates that there is no synergy effect between resources, and no edge is established. All resources are treated as network nodes, and gain edges and hindrance edges are used as weighted edges to connect the nodes, forming a positive and negative differentiated resource interaction and association network. The performance balance baseline module dynamically monitors the overall system performance, statistically analyzes the overall performance mean and fluctuation range based on a sliding time window, and establishes an adaptive performance baseline; it judges whether the current system performance status needs optimization based on the baseline deviation magnitude; The efficiency-first scheduling module, if optimization is required, collects the marginal efficiency contribution and cost consumption of resources, calculates the efficiency return rate, and prioritizes the resources; based on the priority ranking results, it executes resource scheduling under budget constraints.

2. The resource information efficiency evaluation and management system for big data as described in claim 1, characterized in that, The method for obtaining the single resource performance result includes: The constructed resource panorama efficiency vector is used as input, and the functional factors and constraint factors in the resource panorama efficiency vector are normalized. The positive efficiency contribution of the resource is calculated based on the functional factors. The functional factors are quantified by the weighted average method, and the comprehensive positive efficiency value of each functional factor is obtained according to the preset functional factor weight. The negative consumption of resources is calculated based on constraint factors. The constraint factors are quantified through a cost function, and each constraint factor obtains a comprehensive negative consumption value according to the preset constraint factor weight. The positive efficiency contribution and negative consumption are combined to obtain the single resource efficiency value. The calculated single resource efficiency value is recorded as the single resource efficiency result.

3. The resource information efficiency evaluation and management system for big data as described in claim 2, characterized in that, The method for obtaining the overall system performance includes: The single-resource performance result corresponding to each node in the resource interaction association network is used as the initial performance of the node; the performance is propagated for each edge in the resource interaction association network, and the node performance is updated according to the edge weight; The efficiency propagation step is repeated, iterating over all nodes in the resource interaction network until a preset number of iterations is reached; the overall system efficiency is calculated by weighted summation of the efficiency of all nodes in the resource interaction network.

4. The resource information efficiency evaluation and management system for big data as described in claim 3, characterized in that, The method for establishing an adaptive performance baseline includes: The overall system performance is recorded in chronological order to form a system performance time series; the length of the sliding time window is preset, and the overall system performance within the most recent sliding time window is used as the analysis object; within the sliding time window, the mean and fluctuation range of the overall system performance are calculated. Based on the mean and fluctuation range of the overall system performance, the mean of the overall system performance within the sliding time window is used as the center value of the adaptive performance baseline, and the fluctuation range is used as the dynamic fluctuation interval to construct the adaptive performance baseline interval.

5. The resource information efficiency evaluation and management system for big data as described in claim 4, characterized in that, The method for determining whether the current performance status of the system needs optimization based on the baseline deviation includes: A preset baseline deviation threshold is set, and the baseline deviation is calculated based on the average of the overall system performance at the current time point and the overall system performance within the sliding time window. When the baseline deviation exceeds the preset baseline deviation threshold, the system is determined to be in an unsafe or abnormal state and needs to be optimized.

6. The resource information efficiency evaluation and management system for big data as described in claim 5, characterized in that, The method for prioritizing resources includes: If it is determined that the current performance status of the system needs to be optimized, the marginal performance contribution and cost consumption of each resource in the system are collected; the performance return rate of each resource is obtained by calculating the ratio of the marginal performance contribution of each resource to the corresponding cost consumption; and all resources in the system are sorted in descending order according to the performance return rate to form a resource priority list.

7. The resource information efficiency evaluation and management system for big data as described in claim 6, characterized in that, The method for performing resource scheduling under budget constraints includes: Define the total efficiency return of the selected resource set and define the total cost variable. Under the condition that the total cost variable does not exceed the budget constraint, select the optimal resource set to maximize the total efficiency return. Based on the optimal selection result, schedule and allocate the resources.

8. A resource information efficiency evaluation and management method for big data, implemented through any one of claims 1 to 7, characterized in that, include: S1. Factorize the resources to be evaluated, decompose the resource information into functional factors and constraint factors, and form a panoramic efficiency vector of resources. Based on the resource panoramic efficiency vector, the efficiency of a single resource is calculated to obtain the single resource efficiency result; S2. Construct a resource interaction and association network between resources based on the single resource performance results, divide the resource edges into gain edges and hindrance edges, and obtain the overall system performance by propagating the single resource performance in the resource interaction and association network. S3. Dynamically monitor the overall system performance, and establish an adaptive performance baseline by statistically analyzing the overall performance mean and fluctuation range based on the sliding time window. Determine whether the current system performance needs optimization by assessing the baseline deviation magnitude; S4. If optimization is required, collect the marginal performance contribution and cost consumption of resources, calculate the performance return rate, and prioritize the resources. Resource scheduling is performed under budget constraints based on priority ranking results.

Images