Analysis of application software efficiency test strategy design

Basic concepts of efficiency testing

Currently, in the software testing industry, efficiency testing is used to be called performance testing , and the relationship between efficiency testing and performance testing should be the inclusion and inclusion of the relationship. Performance tests measure only time-related indicators and do not measure resources-related indicators. Third-party performance Test is to verify whether the system has reached the development document or user's performance indicators, and also found that the system may exist bottlenecks, to optimize the system.

Performance testing is the evaluation and assessment of response time, transaction rate, and other time-related requirements. The goal of performance evaluation is to verify that performance requirements are met. The purpose of implementing and performing performance reviews is to evaluate and fine-tune the performance behavior of test objects as a function of conditions such as workloads or hardware configurations. Performance testing can be divided into load tests, strength tests, and capacity tests.

A load test is a performance test. The goal is to identify and ensure that the system works well beyond the maximum expected workload. In addition, load tests evaluate performance characteristics. such as response time, transaction rate, and other time-related aspects.

Strength testing is a performance test in which the purpose of implementing and executing such tests is to identify errors that result from insufficient resources or contention for resources. If there is not enough memory or disk space, the test object may exhibit some defects that are not apparent under normal conditions, while others may be caused by contention for shared resources, such as database locks or network bandwidth. Strength testing can also be used to determine the maximum amount of work that a test object can handle.

The capacity test enables the test object to process a large amount of data to determine whether the limit is reached that will cause the software to fail. The capacity test will also determine the maximum load or workload that the Test object can continue to process over a given period of time.

In addition to the monitoring of time, the monitoring of system resources mainly includes processor, memory, network, disk and system.

Implementation strategy of application software efficiency test

At present, in order to meet the needs of different systems, according to the performance test time characteristics and resource characteristics, the application software developed several aspects of efficiency testing strategies.

1. Benchmark Test

Benchmark test refers to the test environment after the validation of the business model for each of the business involved in benchmarking, to obtain the single user performance and resource usage indicators, for multi-user concurrency and mixed scenarios performance test performance analysis to provide a reference.

2. Concurrency test

Concurrent testing refers to a different number of virtual users performing checkpoint operations, which are designed to load tests on checkpoints. After predicting the system response time after the system is put into use under the pressure of a certain user, the performance bottleneck of the system is determined according to the response time, which provides the basis for the optimization and adjustment of the system.

3. Incremental test

Incremental testing refers to a certain period of time (such as 5 seconds, 10 seconds) concurrent different number of virtual users to perform checkpoint operation, the checkpoint to increase user pressure load test, so as to simulate the actual use of the system situation, so that users predict the system after the use of the performance level.

4. Scene test

According to the statistical results of historical data as well as the distribution and use frequency of the system users, simulate different users to perform different operations, such as 10% of users to perform the login operation, 50% of users perform query operations, 40% users perform database update operations, to maximize the simulation of the real scene of the system, To enable users to predict the system into use after the authenticity of the level. Therefore, the system can be optimized and adjusted to avoid the situation that the system does not respond and crashes in the actual situation.

5. Limit test

The limit test is the performance limit test of the function point of the system core function point and the possible pressure bottleneck, that is, the maximum user concurrent number that the function point can withstand. Through the limit test, users can predict whether the system can meet the actual needs after being put into use.

6. Fatigue strength test (stability test)

This test refers to the core function of the system testing the fatigue strength, that is, the maximum number of concurrent users that can be supported under the stable operation of the system, continuous operation time (such as 48 hours), recording the average response time, transaction accuracy, application server and database server CPU utilization, memory usage and other parameters, Review the application server and database server for downtime, memory leaks, and so on. This test usually needs to combine with the scene test, so as to simulate the real environment, the system can maintain the stable running condition under the conditions of long continuous operation.

7. Data storage capacity Test

Using the tools to carry out various tasks continuously, so that the database stores a certain capacity, in the different data storage situation, the implementation of various operations, to obtain system response time. Compared with different data storage capacity, the system response time of different task operation is performed, so that the user can predict the change of the response time of different task operation when the data storage capacity changes in the future.

8. Memory Leak Detection

Using the tools to simulate the system's test points in the single user and multi-user concurrency and long run time, recording the system memory allocation and resource invocation, so as to determine the system test points in the single user and multi-user concurrency when there is a memory leak.

......

This article is reproduced from the 51Testing Software test Network, view the full text: http://www.51testing.com/html/10/n-235510.html