To sum up the performance tests conducted in the past, although the testers have been serious and responsible for the test work from the beginning to the end, after the test report is released, everyone always feels that there is a fly in the ointment, and they have doubts about the test results, especially those with a long time span and for different In the performance comparison test of the test object, there are more or less problems in the following aspects:Inadequate test preparation, unclear test objectives and detailed test plan:

  • 1.nsufficient test preparation, unclear test objectives, and undetailed test plans;The stability and consistency of the test environment are insufficient;

  • 2.Lack of testing and technical reserves for test subjects;

  • 3.Insufficient stability and consistency of the test environment;

Next, we analyze the above problems and discuss how to solve them.

Performance Test Preparation

This is a link that is often overlooked by the testers. After receiving the pressure measurement task, based on the consideration of various other factors, the testers are often eager to progress and immediately devote themselves to the specific test work. Testing, recording, and analysis are busy. It was a great pleasure, but halfway through the work, it was discovered that either the hardware configuration did not meet the requirements, or the network environment was not ideal, or even the software version was wrong, and it was difficult to get off the horse for a while. This was caused by not being ready for testing. So how should we prepare for performance testing?

When doing software projects, there are requirements surveys and analysis, and the same is true when we do testing. After getting the test task, our first task is to analyze the test task. Before starting the test, we must at least clarify the following questions:

  • 1.What will be tested or who will be tested?
  • 2.What question is being tested or what are we trying to figure out or demonstrate?- 3.What factors can affect test results?
  • 4.What kind of test environment is needed?
  • 5.How to test?

Only after carefully investigating the test requirements and carefully analyzing the test tasks, can we clarify the above series of problems. Only when the test tasks are very clear and the test goals are extremely clear, can we formulate a practical test plan.

Clear test objectives, detailed test plan

On the basis of fully understanding the test requirements, it is of great benefit to the implementation of the test to formulate a test plan as detailed as possible. The formulation of the test plan is described in detail in most professional test books, so this article will not repeat it.

Test technical preparation

In the current general environment, it is unrealistic to require testers to master all the software and hardware knowledge in a short time, but usually testers should pay close attention to the research on test tools and test theory. In the test plan, research should be given The test objects and test tools allocate sufficient learning time. Only when we fully grasp the test tools and fully understand the test objects can we implement the test. No matter how hard you try, Jianli’s test on wrong understanding will lead to the opposite result, and it is difficult to prove the problem, let alone use such a test report to convince users.

Configure the test environment

Only on the basis of fully understanding the test object, can we know what kind of configuration each test object requires, and it is possible to configure a relatively fair and reasonable test environment (this is especially important in performance comparison and pressure measurement) .

Taking into account other factors, such as network lock, network speed, display resolution, database permissions, capacity, etc., on the test results. If conditions permit, we’d better configure several different test environments.

Acquisition and processing of test data

In all tests, the collection of test data is relatively difficult, especially for Gis software. Every software has its file format, and some software has several formats. In this case, we can only convert the data in the third-party format into each tested software’s own format. At the same time, the data should also be processed to a certain extent, such as processing data redundancy, processing display styles, etc.

If the data will be updated during the test, the data must be backed up before operation. In addition, the impact of data format and data volume on the test should be evaluated, and multiple sets of data should be prepared if necessary.

Finally, be sure to check the validity of the test data to avoid the impact of corrupted data on the test results.