Software Test Summary (5)

After summing up, the last step of the hierarchical test today is system testing. system testing is part of the Acceptance Work and should be organized and implemented by the user organization. The software development organization should create good conditions for system testing and answer and solve all possible quality problems during testing.

System Testing is performed in a larger scope. In addition to the tested program, the system may also include hardware and other original software. The purpose of system testing is to ensure that the software and other parts of the system work in coordination after the software products are successfully installed in the system and meet the requirements of the Software Requirement Specification.

The test content of the integration system is as follows:

 

Where:

Function Testing mainly checks whether the software functions meet the requirements described in the requirement description. The basic method is to construct some reasonable inputs and check whether the desired outputs are obtained. If the input range is unlimited, the key lies in finding equivalence zones. Another effective test method is the boundary value test, which will be described later.

Performance testing is used to test the Running Performance of software in an integrated system, especially for real-time systems and embedded systems. It is often combined with strength testing.

Security Testing aims to ensure that the protection mechanism installed in the system can protect the system from illegal activities such as illegal intrusion during actual operation, during the test, you need to set up some test cases to try to break through the system's security and confidentiality measures to check whether the system has security and confidentiality vulnerabilities.

The purpose of the recovery test is to ensure that the system can resume normal operation when it is damaged by some external accidents. Various means can be used to forcibly prevent software errors from normal operation, and thus to test the system recovery capability.

In some extreme conditions, the intensity test mainly checks the running status of the software system, for example, some super-large numbers of input data, super-large numbers of users, and super-large numbers of network connections, this is of great significance for understanding the performance, reliability, and robustness of software systems.

Document testing mainly checks the correctness, integrity, and comprehensibility of documents. Correctness means that the software's functions and operations should not be written incorrectly, or the document content should not be in conflict. Integrity means that the document cannot be in the head or end, and the key content cannot be missed; comprehensibility refers to the document content that can be understood and understood by public users.

All stages of the hierarchical test are described here. The next section describes some of the content in the black box test of the classic method of software testing, and finally introduces the criteria for terminating the test.

There are two criteria for the software to terminate the test. One is to define the test strategy and the target to be met. For example, in the white box test, full coverage can be defined as the standard, that is to say, the statement coverage and the branch coverage must reach 100%. If these conditions are met, the test can be stopped. In the black box test, various methods can be specified to design the test case, the test can be stopped when all test cases are used up. Another criterion is to specify at least the number of errors to be identified. For example, to estimate the number of defects in the system based on previous testing experience, the goal is to eliminate 95% of design defects and 98% of coding and structural defects, you can stop the error.