Performance testing and Learning (I)-basic knowledge points

Recently I have been studying performance tests and sorted out some knowledge points.

First, we start with the iso9216 quality model, which shows the dimensions to consider when evaluating a software. As shown in, it is generally evaluated from the six dimensions of functionality, reliability, usability, efficiency, maintainability, and portability. The performance test of software is to evaluate the software from the efficiency dimension.

Briefly describe the content in the efficiency dimension:

• Time features: is the speed of software, can it meet user needs?
• Resource Utilization: after the time feature is reached, is the resource utilization under user expectations?

 

Then, define the performance test:

Observe the system inIn the given environment and scenarioWhether the performance of the system is consistent with the expected goal, judge whether the system has performance defects, and identify the performance bottleneck Based on the test results to improve the system performance.

In the definition, you need to pay special attention to the red part. When describing performance tests, you need to use the given environment and scenario. The given environment involves hardware servers, operating systems, networks, databases, web servers, and application servers. Scenarios are used by users.

 

There are many goals for performance testing, and different testing strategies will be used for different goals.

• Load Testing

In a certain software, hardware, and network environment, performance problems in the system are discovered by changing the load mode and increasing the load.

This test is mainly used to determine the maximum number of users (the system does not crash) and the optimal number of users (the best user experience and the maximum number of users supported by the system) that the system can carry ). The system response time and server resource utilization for different users (common users.

• Stress Testing

In a certain hardware and software environment and network environment, by simulating a large number of virtual users to generate loads on the server, the server's resources are in the limit (beyond the capability range) for a long period of continuous operation.

Objective To test whether the server can work stably under high loads and find the weakest link in the system. Then you can perform positive optimization.

• Concurrency Test

Simulate concurrent accesses to the same application, the same stored procedure or data records, and other concurrent operations by multiple users.

Test whether there are deadlocks, data errors, and other faults. Concurrent resource snatching is prone to problems.

• Stability Test

Simulate a certain number of virtual users to run one or more services in a certain software, hardware, and network environment for a long period of time (7x24.

The purpose is to check whether the system's stability and performance indicators meet expectations during a long period of operation. Resource leakage can be found.

 

Performance testing and Learning (I)-basic knowledge points