A dynamic measurement method and system for complex system simulation resource satisfaction based on multi-task scenarios

Abstract

The present invention proposes a method and system for dynamically measuring the satisfaction of complex system simulation resources based on multi-task scenarios, which belongs to the technical field of complex system simulation and system evaluation. In view of the fact that the existing satisfaction evaluation methods are essentially static evaluation methods based on demand prediction, which cannot meet the demand for dynamic measurement of system simulation resource satisfaction, the present invention decomposes the mission of the simulation object into multiple scenario assumptions, analyzes and obtains ideal resources and minimum resource requirements, and then dynamically monitors the adaptation and satisfaction of the simulation resource situation, supports the measurement of the degree of satisfaction of the simulation task requirements by the resource reserve of the system simulation system, can provide a basis for measuring and optimizing the complex system simulation system, and has practical significance and value for promoting the standardization and normalization of the system simulation industry.

Description

Technical Field

[0001] The present invention belongs to the technical field of complex system simulation and system evaluation, and in particular relates to a method and system for dynamically measuring resource satisfaction of complex system simulation based on a multi-task scenario. Background Art

[0002] Complex system simulation is becoming a high-tech means related to national economy and people's livelihood, national security and national defense construction. Among them, sufficient simulation resources are one of the important aspects to solve system simulation problems.

[0003] At present, there is no feasible method to evaluate the resource satisfaction of complex system simulation systems. It is impossible to scientifically measure the degree to which the simulation resources of the system simulation system meet the simulation needs, and then compare the advantages and disadvantages of different system simulation systems. This has, to a certain extent, caused obvious constraints on the construction of complex system simulation systems and the in-depth application of system simulation technology. Summary of the invention

[0004] Purpose of the invention: To propose a method and system for dynamically measuring the resource satisfaction of complex system simulation based on multi-task scenarios to solve the above-mentioned problems existing in the prior art. Starting from the mission of the simulation object, multi-scenario assumptions are formed, and the ideal resources and minimum resource requirements are obtained through analysis. Then, by dynamically monitoring the adaptation and satisfaction of the simulation resources, the resource reserve of the system simulation system is measured to determine the degree to which it satisfies the simulation task requirements. This can provide a basis for measuring and optimizing the complex system simulation system, and has practical significance and value for promoting the standardization and normalization of the system simulation industry.

[0005] Technical solution: First, a dynamic measurement method for complex system simulation resource satisfaction based on multi-task scenarios is proposed, including the following steps:

[0006] Step 1: Decompose the mission tasks according to the simulation object requirements;

[0007] Step 2: Map the decomposed mission tasks into different mission scenarios;

[0008] Step 3: Perform loading and measurement calculations for different task scenarios in turn, further including the following steps:

[0009] Step 3.1: In the current task scenario, perform measurement demand calculation from four levels: data, model, computing power, and platform;

[0010] Step 3.2, combining the calculation results of step 3.1 with the expert configuration management to generate a simulation resource requirement list for the current scenario;

[0011] Step 3.3, start the simulation operation, and drive the simulation model scheduling and complex system restoration process solution;

[0012] Step 3.4: During the simulation operation, dynamically adapt the simulation resources based on the task scenario, and perform statistical analysis on the active simulation resources to generate a simulation resource utilization list;

[0013] Step 3.5: Based on the simulation resource requirements and actual utilization, dynamic measurement calculation is performed to obtain the measurement indicator sequence data and average value under the current task scenario;

[0014] Step 3.6: As the simulation clock advances, sampling is performed according to a certain simulation step length to monitor the simulation resource satisfaction;

[0015] Step 3.7, determine the simulation result conditions, and terminate the simulation when the end conditions are met;

[0016] Step 4: Perform comprehensive analysis on resource satisfaction data based on the measurement calculation results;

[0017] Step 5: Generate a simulation resource satisfaction map based on the comprehensive analysis results and present it to the user end.

[0018] In some possible implementations of the first aspect, during the dynamic measurement calculation performed in step 3.5, a simulation resource satisfaction measurement model is constructed based on the utility function method, and is calculated according to the two indicators of effectiveness and moderation based on the characteristics of the simulation resources; wherein the measurement indicators of the resource satisfaction measurement model include: data resource abundance, model resource coverage, computing power resource adequacy and platform resource effectiveness.

[0019] The demand satisfaction expression according to the benefit index is:

[0020]

[0021] In the formula, c represents the actual situation of simulation resources; c hig represents the ideal value of simulation resources; c low Indicates the minimum requirement for simulation array resources.

[0022] The demand satisfaction expression according to the moderate index is:

[0023]

[0024] In the formula, c represents the actual situation of simulation resources; c high represents the ideal value of simulation resources; c low Indicates the minimum requirement for simulation resources.

[0025] According to the two calculation methods of efficiency and moderate indicators, the measurement demand values ​​at different levels are quantified, and on the basis of the quantification of the underlying indicators, the comprehensive measurement value of simulation resource satisfaction is solved based on the parallel relationship aggregation method.

[0026] The calculation process of the comprehensive measurement value is:

[0027] S s =1-(1-S d )×(1-S m )×(1-S o )×(1-S p )

[0028] In the formula, S s Indicates the simulation resource satisfaction; S d Indicates the abundance of data resources; S m Indicates model resource coverage; S o Indicates the adequacy of computing resources; S p Indicates the availability of platform resources.

[0029] The quantification process includes: quantifying the abundance of data resources and computing power resources according to moderate indicators; and quantifying the model resource coverage and platform resource effectiveness according to utility indicators.

[0030] Data resource adequacy includes: entity data satisfaction, collaborative data satisfaction and data resource redundancy. Based on the three indicators, the weighted multiplication method is used to evaluate and aggregate the data resource adequacy indicators.

[0031] In some implementations of the first aspect, model resource coverage includes: entity model coverage, task model coverage and model resource value-added, and based on the three indicators included, a weighted multiplication method is used for evaluation aggregation. Computing power resource sufficiency includes: algorithm component sufficiency and computing equipment sufficiency, which are used to measure the sufficiency of computing power resources to support the task scenario restoration mapping requirements, and a weighted summation method is used to obtain a comprehensive quantitative value. Platform resource effectiveness includes: network access effectiveness, storage status effectiveness and operation support effectiveness, which are used to measure the effectiveness of platform resources to support the task scenario restoration mapping requirements, and a weighted summation method is used to obtain a comprehensive quantitative value.

[0032] Among them, the calculation expression of computing power resource sufficiency is:

[0033] C adeq =λ a *EAlg adeq+λ e *EEqu adeq

[0034] In the formula, Cadeq Indicates the adequacy of computing resources; EAlg adeq indicates the mean adequacy of algorithm components under multiple task scenarios within the simulation time; EEqu adeq represents the mean value of computing equipment sufficiency under multiple task scenarios within the simulation duration; a represents the sufficiency weight of the algorithm component; e The weight representing the adequacy of the computing equipment component;

[0035] The calculation expression of platform resource effectiveness is:

[0036] P valid =λ n *ENet valid +λ s *EStr valid +λ r *ERun valid

[0037] Where P valid Indicates the effectiveness of platform resources; ENet valid Indicates the mean value of network access effectiveness under multiple task scenarios during the simulation time; EStr valid represents the mean value of storage state validity under multiple task scenarios within the simulation time; λ n represents the network access effectiveness weight; s Represents the storage state validity weight; λ r The weight representing the effectiveness of operational support.

[0038] Secondly, a dynamic measurement system for complex system simulation resource satisfaction based on multi-task scenarios is proposed to implement a dynamic measurement method for satisfaction. The system includes the following modules: task decomposition module, mapping module, calculation module, analysis module and visualization module.

[0039] Among them, the task decomposition module is used to decompose the mission tasks according to the analysis requirements of the simulation object; the mapping module is used to map the mission tasks decomposed by the task decomposition module into different task scenarios; the operation module is used to perform loading operation and measurement calculation on different task scenarios in turn; the analysis module is used to perform comprehensive analysis on the resource satisfaction data based on the results of measurement calculation; the visualization module is used to generate a simulation resource satisfaction map based on the comprehensive analysis results and present it to the user end.

[0040] Beneficial effect: The present invention proposes a dynamic measurement method and system for complex system simulation resource satisfaction based on multi-task scenarios. Starting from the top-level requirements of simulation applications, the method closely maps the adaptation task scenarios around the mission tasks of simulation objects to form ideal and minimum requirements for simulation resources, thus avoiding the problems of unclear requirement boundaries and demand deviation caused by subjective preferences. The method is highly operational and improves the pertinence, applicability and systematicness of simulation resource demand analysis.

[0041] Among them, resource satisfaction is measured based on multi-task scenarios, and sampling statistical analysis methods are used to comprehensively process resource satisfaction measurement data, avoiding measurement deviation problems caused by demand uncertainty and insufficient coverage under single assumptions, and improving the scientific nature of satisfaction measurement calculations and the accuracy of measurement results.

[0042] At the same time, the present invention dynamically measures the simulation resource satisfaction by monitoring the simulation process data, and uses the real-time resource satisfaction graph and statistical mean graph to intuitively show the changes in resource satisfaction, thereby solving the problem of process-oriented dynamic measurement of resource satisfaction and providing an auxiliary means for optimizing simulation resource utilization.

[0043] In addition, the present invention proposes to use data, models, computing power and platforms as the main simulation resources, adopts the utility function method, comprehensively considers ideal requirements, minimum requirements and actual simulation resource conditions, and performs quantitative calculation of resource satisfaction, thus solving the problem of quantitative evaluation of system simulation resource satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The data processing flow chart of the embodiment of the present invention.

[0045] Figure 2 It is a principle framework diagram of measuring the system simulation resource satisfaction in an embodiment of the present invention.

[0046] Figure 3 Schematic diagram of resource satisfaction measurement indicator architecture in an embodiment of the present invention.

[0047] Figure 4 The following is a flow chart of resource satisfaction measurement in an embodiment of the present invention.

[0048] Figure 5 A schematic diagram of a simulation resource requirement generation framework for strategic needs in an embodiment of the present invention.

[0049] Figure 6 The figure is a schematic diagram of the resource satisfaction measurement calculation process in an embodiment of the present invention.

[0050] Figure 7 Schematic diagram of the main interface of the dynamic measurement system for simulation resource satisfaction in an embodiment of the present invention.

[0051] Figure 8 This is a data resource sufficiency detection diagram in an embodiment of the present invention.

[0052] Fig. 9 This is a model resource coverage map detection map in an embodiment of the present invention.

[0053] Fig.10 This is a graph for detecting the sufficiency of computing resources in an embodiment of the present invention.

[0054] Fig.11 This is a diagram for monitoring the effectiveness of platform resources in an embodiment of the present invention.

[0055] Fig.12 This is a statistical diagram of data resource abundance in an embodiment of the present invention.

[0056] Fig.13 This is a statistical diagram of the model resource coverage in an embodiment of the present invention.

[0057] Fig.14 This is a statistical diagram of computing power resource sufficiency in an embodiment of the present invention.

[0058] Fig.15 This is a statistical diagram of the effectiveness of operating resources in an embodiment of the present invention.

[0059] Fig.16 This is a diagram of the resource satisfaction measurement results of the regional defense system simulation in an embodiment of the present invention.

[0060] Fig.17 This is a comprehensive evaluation diagram of the simulation resource satisfaction of the regional defense system in an embodiment of the present invention.

[0061] Fig.18 This is a mean variance diagram of resource satisfaction of the regional defense system simulation in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In the following description, a large number of specific details are provided to provide a more thorough understanding of the present invention. However, it is apparent to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known in the art are not described.

[0063] The applicant believes that there is little specialized research on the capabilities and effects of complex system simulation systems, let alone specialized research on simulation resource satisfaction. The combat capability satisfaction assessment method and technology in the field of equipment system demonstration and evaluation mainly decomposes and maps from military needs to tasks and then to capabilities, and then aggregates the satisfaction of military needs from the satisfaction of the needs of single capability indicators. This satisfaction assessment method is essentially a static assessment method based on demand forecasting. In actual application, the main problem object is dynamic system simulation, which has dynamic measurement requirements. Therefore, the system capability satisfaction assessment method cannot be copied and used for simulation resource satisfaction assessment.

[0064] Therefore, the present invention proposes a dynamic measurement method and system for complex system simulation resource satisfaction based on multi-task scenarios, which decomposes the mission of the simulation object into multi-scenario assumptions, analyzes the ideal resources and minimum resource requirements, and then dynamically monitors the adaptation and satisfaction of the simulation resources. It supports the measurement of the degree to which the resource reserves of the system simulation system meet the simulation task requirements, which can provide a basis for measuring and optimizing the complex system simulation system, and has practical significance and value for promoting the standardization and normalization of the system simulation industry.

[0065] Embodiment 1

[0066] In one embodiment, simulation resource satisfaction refers to the degree of satisfaction of resources such as models, data, and platforms required to achieve high-fidelity restoration and high-efficiency mapping of complex systems in the physical domain. Among them, dynamic measurement is a process-oriented quantitative evaluation of measurement indicators for the measurement object. Simulation resources are an important guarantee to ensure that the system simulation system is "capable of simulation". Simulation resource satisfaction is to measure the degree to which the simulation system satisfies the simulation task requirements from the resource dimension. At present, there is no technical approach for measuring system simulation resource satisfaction, and it is impossible to scientifically measure the degree to which the simulation resources of the system simulation system satisfy the simulation requirements, and then compare the advantages and disadvantages of different system simulation systems. The existing satisfaction evaluation method is essentially a static evaluation method based on demand prediction, which cannot meet the dynamic measurement requirements of system simulation resource satisfaction. In order to solve the above technical problems, this embodiment proposes a dynamic measurement method for complex system simulation resource satisfaction based on multi-task scenarios. The dynamic measurement method for complex system simulation resource satisfaction based on multi-task scenarios decomposes the mission tasks of the simulation object to form multi-scenario assumptions, analyzes to obtain ideal resources and minimum resource requirements, and then dynamically monitors the adaptation and satisfaction of the simulation resources to achieve dynamic solution of basic measurement indicators, and then aggregates to obtain comprehensive measurement results and conclusions of simulation resource satisfaction.

[0067] like Figure 1 As shown, a dynamic measurement method for complex system simulation resource satisfaction based on a multi-task scenario includes the following steps:

[0068] Step 1: Decompose the mission tasks according to the simulation object requirements;

[0069] Specifically, based on the actual principle of simulation resource satisfaction measurement and according to the system positioning of the system simulation, the simulation tasks involved in the actual application are obtained, and a simulation task list is constructed according to the obtained simulation tasks.

[0070] Step 2: Map the decomposed mission tasks into different mission scenarios;

[0071] Specifically, based on the simulation task list constructed in step 1, the simulation tasks and simulation resources are associated and mapped according to the task scenarios, and a simulation resource measurement indicator system is generated.

[0072] Step 3: Perform loading and measurement calculations for different task scenarios in turn;

[0073] Specifically, first, based on the current task scenario, measurement requirements are calculated from four aspects: data, model, computing power, and platform. Combined with expert configuration management, a list of simulation resource requirements corresponding to the current scenario is generated; secondly, the simulation is started to drive simulation model scheduling and complex system restoration process solution, etc.; thirdly, during the simulation, simulation resources are dynamically adapted based on the task scenario, and statistical analysis is performed on active simulation resources to form a simulation resource utilization list; thirdly, based on simulation resource requirements and actual utilization, dynamic measurement calculations are performed to obtain measurement indicator sequence data and average values ​​under the current task scenario; finally, as the simulation clock advances, sampling is performed according to a certain simulation step size to monitor simulation resource satisfaction, including data demand matching, model resource coverage, operator resource satisfaction, and platform resource adaptation, and the simulation is terminated in a timely manner as needed.

[0074] Step 4: Perform comprehensive analysis on resource satisfaction data based on the measurement calculation results;

[0075] Step 5: Generate a simulation resource satisfaction map based on the comprehensive analysis results and present it to the user end.

[0076] In a preferred embodiment, during the execution of the dynamic measurement method, Figure 2 As shown in the figure, firstly, according to the positioning of the system simulation system, the simulation tasks that may be involved are analyzed, and a simulation task list is constructed. Then, the simulation tasks are associated with the simulation resources through the task scenarios to generate a simulation resource measurement index system. Furthermore, through the task scenario-simulation resource correlation analysis, the simulation task weight is calculated, and the maximum value (i.e., ideal value) and minimum value of the simulation resource demand are set according to the measurement index system. Finally, the simulation resource demand satisfaction of each sub-task and the comprehensive simulation resource demand satisfaction are calculated, and the simulation resource satisfaction measurement conclusion is given.

[0077] Among them, when performing the measurability analysis of simulation resource satisfaction, the corresponding measurement indicators are shown in Table 1, the generated simulation resource demand list is shown in Table 2, and after starting the simulation, the generated simulation resource utilization list is shown in Table 3. In the process of finally performing the dynamic measurement of simulation resource satisfaction, the user's visual interface is as follows Figure 7 As shown, Figure (a) is the data adaptation interface, Figure (b) is the measurement calculation interface, and Figure (c) is the measurement result visualization interface.

[0078] Table 1 Simulation resource satisfaction testability analysis

[0079]

[0080] Table 2 Contents of simulation resource requirement list

[0081]

[0082]

[0083] Table 3 Contents of simulation resource utilization list

[0084]

[0085] According to the needs of actual applications, this embodiment provides a basis for measuring and optimizing complex system simulation systems through the proposed resource satisfaction dynamic measurement method, and further provides technical support for the standardization and regularization of system simulation dynamic measurement in the industry field.

[0086] Embodiment 2

[0087] In a further embodiment based on the first embodiment, in the process of measuring and calculating the system simulation resource satisfaction, a simulation resource satisfaction measurement model is constructed based on the utility function method, and the calculation is performed according to the two indicators of effectiveness and moderation based on the characteristics of the simulation resources. The measurement indicator model is represented as a utility function between the actual situation of the simulation resources and the simulation resource demand, and the value range is [0, 1].

[0088] Among them, the demand satisfaction expression of benefit-based indicators is:

[0089]

[0090] In the formula, c represents the actual situation of simulation resources; c high represents the ideal value of simulation resources; c low Indicates the minimum requirement for simulation resources.

[0091] The demand satisfaction expression of the moderate indicator is:

[0092]

[0093] In the formula, c represents the actual situation of simulation resources; c high represents the ideal value of simulation resources; c low Indicates the minimum requirement for simulation resources.

[0094] In a further embodiment, based on the calculation method of demand satisfaction, the data resource abundance and computing power resource adequacy are quantified according to moderate indicators, and the model resource coverage and platform resource effectiveness are quantified according to utility indicators.

[0095] Based on the quantitative results and the quantification of the underlying indicators, the comprehensive measurement value of the simulation resource satisfaction is solved based on the parallel relationship aggregation method. The corresponding expression is:

[0096] S s =1-(1-S d )×(1-S m )×(1-S o )×(1-S p )

[0097] In the formula, S s Indicates the simulation resource satisfaction; S d Indicates the abundance of data resources; S m Indicates model resource coverage; S o Indicates the adequacy of computing resources; S p Indicates the availability of platform resources.

[0098] In a further embodiment, resource satisfaction measurement indicators include: data resource abundance, model resource coverage, computing power resource adequacy and platform resource effectiveness.

[0099] Among them, data resource sufficiency mainly calculates two types of indicators: data resource satisfaction and data resource redundancy. Data resource satisfaction can be further divided into entity data satisfaction and collaborative data satisfaction. Based on the calculation of the three indicators, the weighted multiplication method is used to evaluate and aggregate the data resource sufficiency indicators. The corresponding expression is:

[0100]

[0101] Where D adeg Indicates the adequacy of data resources; ESat entity It represents the mean value of entity data satisfaction under multiple task scenarios during the simulation time; ESat coor represents the mean value of collaborative data satisfaction under multiple task scenarios within the simulation duration; EDre represents the mean value of data resource redundancy under multiple task scenarios within the simulation duration; λ1 , 2 , 3 Indicates the indicator weight, the default value is 1.

[0102] Entity data resources are characterized by quantity and scale. Entity data satisfaction calculation requires collecting the actual value, ideal value and minimum value of relevant data, and then calculating entity data satisfaction based on the utility function.

[0103] The process of obtaining the actual value of entity data resources during simulation operation includes: ① obtaining static attribute data by counting the number of static data reads of the assumed entity in the database; ② obtaining state parameter data by counting the number of state parameters with write operations; ③ obtaining behavior rule data by counting the number of behavior rules with read operations. For the acquisition of ideal and minimum values ​​of entity data resources, configuration management is performed by experts in the measurement demand configuration link.

[0104] Collaborative data resources are characterized by quantity and scale. Collaborative data satisfaction calculation requires collecting the actual value, ideal value and minimum value of relevant collaborative data, and then calculating collaborative data satisfaction based on the effect function.

[0105] The process of obtaining the actual value of collaborative data resources during simulation operation includes: ① Obtaining task attribute data by counting the number of reads of assumed task attribute data in the database; ② Obtaining interaction parameter data by counting the number of interaction parameters with write operations; ③ Obtaining collaboration rule data by counting the number of collaboration rules with read operations. For the acquisition of ideal and minimum values ​​of collaborative data resources, configuration management is performed by experts in the measurement demand configuration link.

[0106] Data resource redundancy refers to the degree to which data resource conditions exceed expectations. It is measured by the ratio of the difference between the actual number of data resources and the expected number of data resources to the expected number of data resources. The corresponding expression is:

[0107] Dre=(NumData t -NumData high ) / NumData high

[0108]

[0109] In the formula, Dre represents the data resource redundancy; NumData t Indicates the actual number of data resources; NumData high Indicates the expected number of data resources, that is, the ideal number of data resources.

[0110] Model resource coverage mainly calculates two types of indicators: model resource coverage and model resource value-added. Among them, model resource coverage is divided into entity model coverage and task model coverage. Model resource coverage calculation is to evaluate and aggregate entity model coverage, task model coverage and model resource value-added, using the weighted product method, and the corresponding expression is:

[0111]

[0112] Where M cov Indicates model resource coverage; ECov entity Represents the mean coverage of the entity model under multiple task scenarios within the simulation time; ECov task Indicates the mean coverage of the task model under multiple task scenarios within the simulation time; EInc m represents the mean value of model resource appreciation under multiple task scenarios within the simulation duration; e , t , m Indicates the indicator weight, the default value is 1.

[0113] The entity model coverage is measured and calculated from two aspects: model type and model quantity. The comprehensive quantitative result of entity model coverage is obtained by weighted multiplication method. The corresponding expression is:

[0114]

[0115] In the formula, Cov entity Indicates entity model coverage; Ctpye entity Indicates entity model type coverage; Snum entity Indicates the quantity satisfaction of the entity model; t represents the entity model type coverage weight; n Indicates the weight of the entity model quantity satisfaction. The default value of the weight is 1.

[0116] The entity model type coverage refers to the degree of matching between the entity model types of the system simulation system and the entity model types of the task scenario. It is characterized by the ratio of the number of matching entity types to the number of entity types in the task scenario. The corresponding expression is:

[0117] Ctpye entity =Snum entity / Cnum entity

[0118] In the formula, Snum entity Indicates the number of entity model types of the system simulation system that can match the entity types of the task scenario; Cnum entity Indicates the number of entity types measured in the task scenario.

[0119] The satisfaction degree of the number of entity models refers to the degree to which the number of entity models of the system simulation system satisfies the number of entity models expected in the task scenario. It is a benefit-type indicator and is quantified using the utility function method.

[0120] The calculation process of task model coverage is the same as that of entity model coverage, and the corresponding calculation expression is:

[0121]

[0122] In the formula, Cov task is the task model coverage; Ctpye task is the task model type coverage; Snum task is the satisfaction degree of the number of task models; α represents the task model type coverage weight; β represents the weight of the satisfaction degree of the number of task models, and the default value of the weight is 1.

[0123] Task model type coverage refers to the degree of matching between the task model type of the system simulation system and the task model type of the task scenario. It is characterized by the ratio of the number of matched task types to the number of task types in the task scenario. The corresponding calculation expression is:

[0124] Ctpye task =Snum task / Cnum task

[0125] In the formula, Snum task Indicates the number of task model types of the system simulation system that can match the task scenario task types; Cnum task Indicates the number of task types measured in the task scenario.

[0126] The satisfaction degree of the number of task models refers to the degree to which the number of task models of the system simulation system satisfies the number of task models expected by the task scenario. It is a benefit-type indicator and is quantified using the utility function method.

[0127] The value-added degree of model resources refers to the degree to which the value of the model resources of the simulation system exceeds the expected model resource requirements, including type value-added and quantity value-added. The corresponding expression is:

[0128]

[0129] In the formula, Inc m Indicates the value-added degree of model resources; Inc type Indicates the value-added degree of the model type; Inc count Indicates the degree of increase in the number of models; ty A measure of the value added of a model type;co A measure of the degree of increase in the number of models; Snum type Indicates the number of types of model resources in the system; Cnum type Indicates the number of model types required based on task scenario measurement, that is, the ideal number of model types; Snum count Indicates the total number of model resources in the system; Cnum count It represents the total number of models required based on the task scenario, that is, the ideal number of models.

[0130] The calculation of computing power resource sufficiency is to measure the sufficiency of computing power resources to support the task scenario restoration mapping requirements. The comprehensive quantitative value of this indicator is obtained by weighted summing the two indicators of algorithm component sufficiency and computing equipment sufficiency. The corresponding expression is:

[0131] C adeq =λ a *EAlg adeq+λ e *EEqu adeq

[0132] In the formula, C adeq Indicates the adequacy of computing resources; EAlg adeq indicates the mean adequacy of algorithm components under multiple task scenarios within the simulation time; EEqu adeq represents the mean value of computing equipment sufficiency under multiple task scenarios within the simulation duration; a represents the sufficiency weight of the algorithm component; e The weight representing the sufficiency of computing equipment components.

[0133] The adequacy of algorithm components is a benefit-based indicator. The minimum number of algorithm components is taken as the basic demand, and the ideal demand of algorithm components is taken as the expected demand. The closer the number of algorithm components is to the ideal demand, the more sufficient the algorithm component resources are. When the algorithm components exceed the ideal demand, the adequacy is set to 1. The algorithm component adequacy indicator is quantitatively calculated using a benefit-based effect function.

[0134] The adequacy of computing equipment is a benefit-based indicator. The minimum number of computing equipment is taken as the basic demand, and the ideal demand for computing equipment is taken as the expected demand. The closer the number of computing equipment is to the ideal demand, the more sufficient the computing equipment resources are. When the number of computing equipment exceeds the ideal demand, the adequacy is set to 1. The computing equipment adequacy indicator is quantified using a benefit-based effect function.

[0135] The platform resource effectiveness measures the effectiveness of platform resources that support the task scenario restoration mapping requirements. The comprehensive quantitative value of this indicator is obtained by weighted summation of network access effectiveness, storage status effectiveness, and operation support effectiveness. The corresponding expression is:

[0136] P valid =λ n *ENet valid +λ s *EStr valid +λ r *ERun valid

[0137] Where P valid Indicates the effectiveness of platform resources; ENet valid Indicates the mean value of network access effectiveness under multiple task scenarios during the simulation time; EStr valid represents the mean value of storage state validity under multiple task scenarios within the simulation time; λ n represents the network access effectiveness weight; s Represents the storage state validity weight; λ r The weight representing the effectiveness of operational support.

[0138] The network access effectiveness index is measured by the ratio of the number of effective network accesses to the total number of accesses. The corresponding expression is:

[0139] Net valid =NetNum vin / NetNum in

[0140] Where NetNum vin Indicates the number of valid network accesses; NetNum in Indicates the total number of network accesses.

[0141] The storage state effectiveness index is measured by the ratio of the number of effective storage times to the total number of storage times. The corresponding expression is:

[0142] Str valid =StrNum vst / StrNum st

[0143] In the formula, StrNum vst Indicates the number of valid network accesses; StrNum st Indicates the total number of network accesses.

[0144] The operation support effectiveness index is measured by the ratio of the number of effective operation supports to the total number of operation support services during the simulation process. The corresponding expression is:

[0145] Run valid =RunNum vru / RunNum ru

[0146] Where, RunNum vru Indicates the number of effective network accesses; RunNum ru Indicates the total number of network accesses.

[0147] In a preferred embodiment, as the simulation clock advances, sampling is performed at a certain simulation step length to monitor the simulation resource satisfaction, including data demand matching, model resource coverage, operator resource satisfaction, and platform resource adaptation. Fig.15 As shown, the simulation is terminated in due time as needed. When all task scenarios are measured, the output resource satisfaction data is comprehensively analyzed to form a simulation resource satisfaction map to display the measurement results, as shown in Fig.16 , 17 , 1 8. Among them, the mean represents the overall satisfaction of the simulation resources, and the variance represents the deviation of the resource satisfaction. It can be seen from the results of the embodiment that the satisfaction is 89.9%, and the deviation is small.

[0148] Embodiment 3

[0149] In one embodiment, a complex system simulation resource satisfaction dynamic measurement system based on a multi-task scenario is proposed to implement a satisfaction dynamic measurement method. The system includes: a task decomposition module, a mapping module, a calculation module, an analysis module and a visualization module.

[0150] In a further embodiment, according to the analysis requirements of the simulation object, a task decomposition module is used to decompose the mission task, and then based on the decomposed mission task, a mapping module is used to complete the mapping of the mission task and different task scenarios; then, an operation module is used to perform loading operation and measurement calculation on different task scenarios in turn, and the analysis module is used to perform a comprehensive analysis on the results of the operation module; finally, the comprehensive analysis results are used to generate a simulation resource satisfaction map, and presented to the user end through a visualization module.

[0151] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes may be made to it in form and detail without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A dynamic measurement method for resource satisfaction of complex system simulation based on multi-task scenario, characterized in that: The following steps are involved: Step 1: Decompose the mission tasks according to the simulation object requirements; Step 2: Map the decomposed mission tasks into different mission scenarios; Step 3, sequentially perform loading operation and measurement calculation for typical scenarios under different task scenarios; in the process of executing the measurement calculation, a dynamic measurement calculation mode is adopted, combined with simulation resource requirements and actual utilization, based on the utility function method, a simulation resource satisfaction measurement model is constructed, and according to the characteristics of the simulation resources, calculation is performed according to the two indicators of efficiency and moderateness; The measurement indicators of the resource satisfaction measurement model include: data resource sufficiency, model resource coverage, computing power resource sufficiency and platform resource effectiveness; The data resource adequacy includes: entity data satisfaction, collaborative data satisfaction and data resource redundancy. Based on the three indicators, a weighted multiplication method is used to evaluate and aggregate the data resource adequacy indicators. The model resource coverage includes: entity model coverage, task model coverage and model resource value-added, and the weighted multiplication method is used to evaluate and aggregate the three indicators. The computing power resource adequacy includes: algorithm component adequacy and computing equipment adequacy, which are used to measure the adequacy of computing power resources to support the task scenario restoration mapping requirements, and a comprehensive quantitative value is obtained by weighted summation. The calculation expression of the computing power resource adequacy is: C adeq =λ a *EA lgadeq+λ e * Eequ adeq In the formula, C adeq Indicates the adequacy of computing resources; EAlgadeq indicates the average adequacy of algorithm components under multiple task scenarios within the simulation time; EEqu adeq represents the mean value of computing equipment sufficiency under multiple task scenarios within the simulation duration; a represents the sufficiency weight of the algorithm component; e The weight representing the adequacy of the computing equipment component; The platform resource effectiveness includes: network access effectiveness, storage status effectiveness and operation support effectiveness, which is used to measure the effectiveness of platform resources supporting the task scenario restoration mapping requirements, and a comprehensive quantitative value is obtained by weighted summation. The calculation expression of the platform resource effectiveness is: P valid =λ n *Enet valid +λ s *Estr valid +λ r *Erun valid Where P valid Indicates the effectiveness of platform resources; ENet valid Indicates the mean value of network access effectiveness under multiple task scenarios during the simulation time; EStr valid represents the mean value of storage state validity under multiple task scenarios within the simulation time; λ n represents the network access effectiveness weight; s Represents the storage state validity weight; λ r The weight representing the effectiveness of operational support; Step 4: Perform comprehensive analysis on resource satisfaction data based on the measurement calculation results; Step 5: Generate a simulation resource satisfaction map based on the comprehensive analysis results and present it to the user end.

2. According to claim 1, a method for dynamically measuring resource satisfaction of complex system simulation based on multi-task scenarios is characterized in that: The process of executing the measurement calculation in step 3 includes the following steps: Step 3.1: In the current task scenario, perform measurement demand calculation from four levels: data, model, computing power, and platform; Step 3.2, combining the calculation results of step 3.1 with the expert configuration management to generate a simulation resource requirement list for the current scenario; Step 3.3, start the simulation operation, and drive the simulation model scheduling and complex system restoration process solution; Step 3.4: During the simulation operation, dynamically adapt the simulation resources based on the task scenario, and perform statistical analysis on the active simulation resources to generate a simulation resource utilization list; Step 3.5: Based on the simulation resource requirements and actual utilization, dynamic measurement calculation is performed to obtain the measurement indicator sequence data and average value under the current task scenario; Step 3.6: As the simulation clock advances, sampling is performed according to a certain simulation step length to monitor the simulation resource satisfaction; Step 3.7: Determine the simulation result conditions, and terminate the simulation when the end conditions are met.

3. According to claim 1, a method for dynamically measuring resource satisfaction of complex system simulation based on multi-task scenarios is characterized in that: The demand satisfaction expression according to the benefit index is: In the formula, c represents the actual situation of simulation resources; c high represents the ideal value of simulation resources; c low Indicates the minimum requirement for simulation array resources.

4. According to the method of claim 1, the method is characterized in that: The demand satisfaction expression according to the moderate index is: In the formula, c represents the actual situation of simulation resources; c high represents the ideal value of simulation resources; c low Indicates the minimum requirement for simulation resources.

5. The method for dynamically measuring resource satisfaction of complex system simulation based on multi-task scenarios according to claim 1 is characterized in that: According to the two indicators of effectiveness and moderate calculation methods, the measurement demand values ​​at different levels are quantified, and based on the quantification of the underlying indicators, the comprehensive measurement value of simulation resource satisfaction is solved based on the parallel relationship aggregation method; The quantification process includes: Quantify the adequacy of data resources and computing resources according to moderate indicators; Quantify the model resource coverage and platform resource effectiveness according to utility indicators.

6. The method for dynamically measuring resource satisfaction of complex system simulation based on multi-task scenarios according to claim 5 is characterized in that: The calculation process of the comprehensive measurement value is: S s =1-(1-S d )×(1-S m )×(1-S o )×(1-S p ) In the formula, S s Indicates the simulation resource satisfaction; S d Indicates the abundance of data resources; S m Indicates model resource coverage; S o Indicates the adequacy of computing resources; S p Indicates the availability of platform resources.

7. A complex system simulation resource satisfaction dynamic measurement system based on a multi-task scenario, used to implement the satisfaction dynamic measurement method according to any one of claims 1 to 6, characterized in that: Includes the following modules: The task decomposition module is configured to decompose the mission task according to the analysis requirements of the simulation object; A mapping module is configured to map the mission tasks decomposed by the task decomposition module into different task scenarios; The computing module is configured to sequentially perform load operation and measurement calculation for different task scenarios; An analysis module configured to perform a comprehensive analysis on the resource satisfaction data based on the results of the measurement calculation; The visualization module is configured to generate a simulation resource satisfaction map based on the comprehensive analysis results and present it to the user end.

Images