Characteristics of software errors.

An important feature of the creation of complex PS and ... is the lack of a reference, which must correspond to the program text and the results of its operation.
During debugging and testing, secondary errors are first detected, i.e. the results of the initial defects, which are the primary errors or causes of the detected anomalies. The manifestations of defects and errors in different degrees affect the performance of the program.
The magnitude of damage manifestations of defects and secondary errors are divided into:
1. failures that do not significantly affect the performance of the program and damage that can be neglected.
2. ordinary failures, the damage from which is within acceptable limits;
3. catastrophic failures, the damage from which is so great that it determines the safety of using this complex of programs.
The characteristics and the specific implementation of the primary errors do not allow to unambiguously predict the types and the degree of manifestation of secondary errors and their influence on the reliability of the PS. In practice, the simplest errors of programs and data can lead to disastrous consequences. At the same time, system defects can slightly degrade performance and do not affect the safety of the programs. Error statistics in software packages and their characteristics can serve as guidelines for developers when allocating debugging efforts. Registering, collecting and analyzing the characteristics of errors in programs is a complex and time-consuming process. Software developers do not advertise errors. All this prevents the receipt of effective error data.
Primary errors in PS in the order of the complexity of their detection can be divided into the following types:
1. technological - these are errors in the preparation of machine media, documentation and errors in the input of programs into the PC memory and their output to display devices;
2. software errors due to incorrect recording of the source text of programs in the programming language and errors in translating programs into object code;
3. algorithmic errors associated with the incomplete formation of the necessary conditions for the solution and incorrectly assigned tasks;
4. system errors caused by the deviation of the PS functioning in a real system and the deviation of the characteristics of external objects from those assumed during the design.
The listed errors vary considerably in frequency and methods of their detection at various stages of program design. With autonomous and first complex debugging, the proportion of system errors is approximately 10%. At the final stages of complex debugging, it increases to 35-40%. In the process of tracking system errors are predominant and make up to 80%. The frequency of occurrence of secondary errors in the operation of the program and the frequency of their detection at failure depend on the total number of primary errors in the program or on the probable error in the team. The number of secondary errors in the program that are detected per unit of time during the testing process is the most accessible for measuring. It is also possible to register failures during the operation of the program.