CN-121996516-A - Capacity calculation method and device based on full-link pressure test

CN121996516ACN 121996516 ACN121996516 ACN 121996516ACN-121996516-A

Abstract

The invention discloses a capacity calculation method and device based on a full-link pressure test, wherein the method comprises the steps of obtaining a transaction link according to transaction service scene analysis, wherein the transaction service scene is used for describing a specific service flow and a system interaction process which are experienced from initiation, processing to completion of transaction, setting up a test environment according to the transaction link, deploying a test transaction model, monitoring an initiation end-to-end transaction test based on the test environment, the transaction link and the test transaction model, monitoring a test execution process to obtain monitoring data, determining a mapping relation between resource utilization rate and throughput capacity of each subsystem according to the monitoring data, and calculating the throughput capacity of each subsystem according to the mapping relation and resource utilization rate available amount of each subsystem in a production environment, wherein the resource utilization rate available amount represents resource amount meeting safe operation of the subsystem. The method and the device can intelligently estimate the capacity of each physical subsystem on the transaction link, and improve the accuracy of capacity estimation.

Inventors

WANG YEHONG
HU FEIYU
YU YANAN
HAN JIUXUE
WANG YANHUA

Assignees

中国建设银行股份有限公司

Dates

Publication Date: 20260508
Application Date: 20251215

Claims (20)

1. The capacity calculation method based on the full link pressure test is characterized by comprising the following steps of: The transaction service scene is used for describing a specific service flow and a system interaction process which are undergone by the transaction from the initiation, the processing and the completion; Setting up a test environment according to the transaction link and deploying a test transaction model, wherein the test environment comprises a test server corresponding to a subsystem in the transaction link; Based on the test environment, the transaction link and the test transaction model, initiating an end-to-end transaction test, monitoring a test execution process to obtain monitoring data, wherein the monitoring data comprises the resource utilization rate and the throughput capacity of each subsystem; and determining a mapping relation between the resource utilization rate and the throughput capacity of each subsystem according to the monitoring data, and calculating the throughput capacity of each subsystem according to the mapping relation and the available resource utilization rate of each subsystem in the production environment, wherein the available resource utilization rate represents the resource quantity meeting the safe operation of the subsystem.
2. The method of claim 1, wherein building a test environment from a transaction link comprises: and (3) referring to the resource use condition of each subsystem in the production environment, and converting and deploying the test server resources according to a preset proportion.
3. The method of claim 1, wherein the plurality of subsystems in the transaction link comprises a master system, wherein the master system is determined according to one of software version changes, importance of executing transactions, discovery risk, or any combination thereof; building a test environment according to a transaction link, including: determining the resource conversion proportion of a test server corresponding to a main test system by referring to the resource use condition of the main test system in the production environment; And determining the resource conversion ratio of other test servers according to the resource conversion ratio of the test server corresponding to the main test system.
4. The method of claim 1, wherein the test server deployment actual physical location is the same as an actual physical location of a corresponding subsystem.
5. The method of claim 1, wherein deploying the test transaction model comprises: Each column in the transaction list comprises transaction names and transaction duty ratios, and the transaction lists are ordered from big to small according to the transaction duty ratio; taking the clauses of the preset duty ratio in front of the transaction list to form a test transaction model of each subsystem; and deploying the test transaction model of each subsystem to a test server corresponding to each subsystem.
6. The method of claim 1, wherein initiating the end-to-end transaction test comprises: The method comprises the steps of generating pressure in a gradient increasing mode based on a test environment, a transaction link and a test transaction model, wherein the gradient increasing mode comprises the steps that the difference of resource utilization rates corresponding to two adjacent gradients reaches more than a specified percentage of the resource utilization rate of the previous gradient, or the difference of throughput capacities corresponding to two adjacent gradients reaches more than a specified percentage of the throughput capacity of the previous gradient.
7. The method of claim 6, wherein initiating the end-to-end transaction test comprises: Monitoring whether the resource utilization rate of each subsystem on a transaction link reaches a preset threshold; And (3) for the subsystem of which the resource utilization rate does not reach the preset threshold value, increasing background compensation pressure until the resource utilization rate reaches the preset threshold value, wherein the background compensation pressure is sent out according to a test transaction model of the subsystem.
8. The method of claim 6, wherein the monitoring data includes resource usage and throughput capacity for each gradient of each subsystem; Determining a mapping relation between the resource utilization rate and the throughput capacity of each subsystem according to the monitoring data, wherein the mapping relation comprises the following steps: And performing linear regression fitting by utilizing the resource utilization rate and the throughput capacity of each gradient to obtain the mapping relation between the resource utilization rate and the throughput capacity of each subsystem.
9. The method of claim 6, wherein determining a mapping between each subsystem resource usage and throughput capacity based on the monitored data comprises: And performing binary once equation fitting by utilizing the resource utilization rate and the throughput capacity of each gradient to obtain the mapping relation between the resource utilization rate and the throughput capacity of each subsystem.
10. The method of claim 1, wherein the resource usage availability is a difference between a maximum resource usage threshold and a server's resource usage when idle.
11. The method of claim 1, further comprising, after deriving throughput capacity of each subsystem from the mapping relationship and the resource usage availability of each subsystem in the production environment: calculating the ratio of the calculated throughput capacity of each subsystem to the throughput capacity of each subsystem in the production environment to obtain the ratio of each subsystem; And determining the subsystem with the resource bottleneck and/or the subsystem with the resource waste according to the size difference of the ratio of the subsystems.
12. The capacity calculation device based on the full link pressure test is characterized by comprising: The transaction system comprises a transaction link analysis module, a transaction service scene analysis module and a transaction processing module, wherein the transaction link analysis module is used for obtaining a transaction link according to transaction service scene analysis, wherein the transaction link is a path formed by a plurality of subsystems through which a transaction passes from initiation to completion; The test preparation module is used for building a test environment according to the transaction link and deploying a test transaction model, wherein the test environment comprises a test server corresponding to the transaction link subsystem; The test monitoring module is used for initiating an end-to-end transaction test based on a test environment, a transaction link and a test transaction model, monitoring a test execution process and obtaining monitoring data, wherein the monitoring data comprises resource utilization rate and throughput capacity of each subsystem; The capacity calculation module is used for determining the mapping relation between the resource utilization rate and the throughput capacity of each subsystem according to the monitoring data, and calculating the throughput capacity of each subsystem according to the mapping relation and the available resource utilization rate of each subsystem in the production environment, wherein the available resource utilization rate represents the resource quantity meeting the safe operation of the subsystem.
13. The apparatus of claim 12, wherein the test preparation module is specifically configured to: Each column in the transaction list comprises transaction names and transaction duty ratios, and the transaction lists are ordered from big to small according to the transaction duty ratio; taking the clauses of the preset duty ratio in front of the transaction list to form a test transaction model of each subsystem; and deploying the test transaction model of each subsystem to a test server corresponding to each subsystem.
14. The apparatus of claim 12, wherein the test monitor module is specifically configured to: The method comprises the steps of generating pressure in a gradient increasing mode based on a test environment, a transaction link and a test transaction model, wherein the gradient increasing mode comprises the steps that the difference of resource utilization rates corresponding to two adjacent gradients reaches more than a specified percentage of the resource utilization rate of the previous gradient, or the difference of throughput capacities corresponding to two adjacent gradients reaches more than a specified percentage of the throughput capacity of the previous gradient.
15. The apparatus of claim 14, wherein the test monitor module is specifically configured to: Monitoring whether the resource utilization rate of each subsystem on a transaction link reaches a preset threshold; And (3) for the subsystem of which the resource utilization rate does not reach the preset threshold value, increasing background compensation pressure until the resource utilization rate reaches the preset threshold value, wherein the background compensation pressure is sent out according to a test transaction model of the subsystem.
16. The apparatus of claim 14, wherein the monitoring data comprises resource usage and throughput capacity for each gradient of each subsystem; The capacity calculation module is specifically used for: And performing linear regression fitting by utilizing the resource utilization rate and the throughput capacity of each gradient to obtain the mapping relation between the resource utilization rate and the throughput capacity of each subsystem.
17. The apparatus of claim 14, wherein the capacity estimation module is specifically configured to: And performing binary once equation fitting by utilizing the resource utilization rate and the throughput capacity of each gradient to obtain the mapping relation between the resource utilization rate and the throughput capacity of each subsystem.
18. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 11 when executing the computer program.
19. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, implements the method of any of claims 1 to 11.
20. A computer program product, characterized in that the computer program product comprises a computer program which, when executed by a processor, implements the method of any of claims 1 to 11.

Description

Capacity calculation method and device based on full-link pressure test Technical Field The invention relates to the technical field of computers, in particular to a capacity calculation method and device based on full-link pressure test. Background The data center is used as an operation unit for safe production and is responsible for guaranteeing the stable operation of important business scenes. The capacity calculation of the production system is a key link of operation management of the data center, and the core meaning of the capacity calculation is derived from the refined operation requirement on resource planning. The accurate capacity calculation supports the production resource planning decision, and is an important basis for balancing the input and output efficiency of resources. The conventional system capacity calculation methods have the characteristics but also have the limitations: Based on the production history data analysis method, the method is only applicable to the put-to-production system. Since the data is completely dependent on the production data, the data sampling near the resource bottleneck is rarely generated in the production environment with stable operation, and the data sampling points are all the homogeneous data with low resource consumption. The sampling data cannot be flexibly manufactured according to the requirements, and although a production pressure measurement way exists, the production pressure measurement generally needs data dyeing and cleaning, the preparation and ending workload investment is high, and the risk factors of safe production are increased. The testing environment pressure test is mainly a single-system test and is mainly used for finding out an internal bottleneck deployment unit of a single system, potential performance problems caused by mutual dependence of systems on an end-to-end transaction link cannot be found out, and the capacity estimation accuracy of each system is poor. Disclosure of Invention The embodiment of the invention provides a capacity estimation method based on a full link pressure test, which is used for intelligently estimating the capacity of each physical subsystem on a transaction link and improving the accuracy of capacity estimation, and comprises the following steps: The transaction service scene is used for describing a specific service flow and a system interaction process which are undergone by the transaction from the initiation, the processing and the completion; Setting up a test environment according to the transaction link and deploying a test transaction model, wherein the test environment comprises a test server corresponding to a subsystem in the transaction link; Based on the test environment, the transaction link and the test transaction model, initiating an end-to-end transaction test, monitoring a test execution process to obtain monitoring data, wherein the monitoring data comprises the resource utilization rate and the throughput capacity of each subsystem; and determining a mapping relation between the resource utilization rate and the throughput capacity of each subsystem according to the monitoring data, and calculating the throughput capacity of each subsystem according to the mapping relation and the available resource utilization rate of each subsystem in the production environment, wherein the available resource utilization rate represents the resource quantity meeting the safe operation of the subsystem. The embodiment of the invention also provides a capacity estimation device based on the full link pressure test, which is used for intelligently estimating the capacity of each physical subsystem on the transaction link and improving the accuracy of capacity estimation, and comprises the following components: The transaction system comprises a transaction link analysis module, a transaction service scene analysis module and a transaction processing module, wherein the transaction link analysis module is used for obtaining a transaction link according to transaction service scene analysis, wherein the transaction link is a path formed by a plurality of subsystems through which a transaction passes from initiation to completion; The test preparation module is used for building a test environment according to the transaction link and deploying a test transaction model, wherein the test environment comprises a test server corresponding to the transaction link subsystem; The test monitoring module is used for initiating an end-to-end transaction test based on a test environment, a transaction link and a test transaction model, monitoring a test execution process and obtaining monitoring data, wherein the monitoring data comprises resource utilization rate and throughput capacity of each subsystem; The capacity calculation module is used for determining the mapping relation between the resource utilization rate and the throughput capacity of each subsystem according to the monitoring data, and ca