A Method of Generating Test Data for Software Reliability

Abstract

The present invention provides a method for generating test data for software reliability. The method includes: a software failure mode analysis step, a software space analysis step, a mapping step and a test data generation step. According to the software failure mode obtained in the early stage of the development stage, the invention analyzes the space of the software, determines the input variables, output variables, the value space of the variables, the value rules and constraints of the variables in the value space, and then The failure mode is mapped to the software space, and finally, based on the mapping function between the failure mode and the software space, test data is generated according to possible input and output values ​​of the software during use. The invention utilizes early information in the development stage, thereby improving the ability of test cases to expose defects and reducing the repetition of test cases.

Description

technical field [0001] The present application relates to the technical field of software testing, in particular to a method for generating test data for software reliability. Background technique [0002] At present, the mainstream technology is the method of generating and executing software reliability test cases based on the operation profile. There are many defects in the implementation process of this test method, mainly including the following aspects: 1) The accuracy of the operation profile is not high: when the software is running, the execution logic is often unpredictable, the usage scenarios are complex and changeable, the data interaction is frequent, It is very difficult to accurately describe the software usage and its probability information for complex features such as difficult to measure operating rules; It was not used in the software reliability testing process, resulting in early software failure information not being effectively transmitted to the te...

AI Technical Summary

Problems solved by technology

There are many defects in the implementation process of this test method, mainly including the following aspects: 1) The accuracy of the operation profile is not high: when the software is running, the execution logic is often unpredictable, the usage scenarios are complex and changeable, the data interaction is frequent, It is very difficult to accurately describe the software usage and its probability information for complex features such as difficult to measure operating rules; It was not used in the software reliability testing process, resulting in early software failure information not being effectively transmitted to the testing stage. Targeted testing was not carried out for software failure modes in the software reliability testing, resulting in information loss, leading to software reliability testing The error rate of use cases is low, and the ability to find software defects is weak; 3) The repetition of test cases is serious: the test cases are obtained by random sampling of software inputs, and the software inputs that are frequently used in practice are tested in the software reliability test. The proportion of use cases is high, resulting in serious duplication of test cases

Therefore, the method of generating and executing software reliability test cases based on the operation profile has the above defects, resulting in poor applicability and applicability, and it is difficult to be widely used in the development of equipment software

Images