11.3. Constructive characteristics of the quality of complex software

Constructive characteristics are divided into two groups: quantitative and qualitative, which differ in the possibilities of concretization of measures and scales. Two groups of standardized PS quality characteristics - Reliability and Efficiency are most accessible to quantitative measurements. For them, Table 11.2 presents examples of possible measures and scales for measuring the main quantitative attributes of sub-characteristics of quality. They can serve as reference points in the selection and determination of the required values ​​of these quality indicators in the PS specifications.

Table 11.2 Basic quantitative specifications software tools And their attributes

Reliability: the properties of the software package to ensure a fairly low probability of loss of performance - failure in the process

the functioning of the PS in real time. The basic attributes of reliability can be objectively measured and compared with the requirements. The requirements for the attribute values ​​of the sub-characteristics of completeness - permissible mean time to failure - are set in the absence of an automatic restart and in the presence of an administrator who controls the performance of the software. By using hardware-software mechanisms for automatic restart, this operating time can be increased with the appearance of failures, i.e. with some failures, it is possible to automatically detect them and ensure operative restoration of working capacity, as a result of which the values ​​of stability and time between failures increase. This should be taken into account when determining the requirements for availability - the probability to find the PS in working condition. Just as in the formation of requirements for correctness (see Table 11.1), for reliability, test coverage in the process of debugging the structure and functions of software components and software as a whole is of great importance.

The reliability of the functioning of programs is a dynamic concept, manifested in time, and differs significantly from the concept of static correctness of programs. Reliability of PS is most fully characterized by stability or ability for reliable operation and recoverability of the operational state after the failures or failures occurred. In turn, sustainability depends on the degree of coverage of the functions and structure of the programs with tests, on the level of unresolved defects and errors (completeness) and on the PS ability to respond to their manifestations so that it does not affect the reliability indicators. The latter are determined by the effectiveness of control data from the external environment and from the detection of anomalies of the functioning of the PS. Under real conditions, for various reasons, the source data may fall into the ranges of values ​​that have not been verified during development and testing, as well as not specified by the requirements of specifications and technical specifications, causing failures and failures. In this case, an incorrect program can function completely reliably.

Completion: the PS property does not fall into failure states due to errors and defects in programs and data. The number or density of manifestations of hidden defects and errors directly reflects

on the duration of the normal functioning of the program between failures. Completion can be characterized by time (duration) to failure (in the absence of automatic recovery - restart), usually measured in hours. This failure is affected only by failures, due to the defects that have appeared. They may be due to incomplete test coverage during testing of components and PS as a whole, as well as insufficient completeness of testing their functions.

Resistance to defects and errors: the property of PS to automatically maintain a given level of quality of functioning during the manifestation of defects and errors or violations of the installed interface. To do this, temporary, software and information redundancy should be introduced in the PS, which realizes the rapid detection of defects and operational errors, their identification and automatic restoration (restart) of the normal functioning of the PS. Effective, prompt elimination of the manifestations of defects, errors and incorrect interaction with the operating and external environment determine the subcharacteristics - the stability of program complexes.

Recoverability: the property of the PS in case of failure to resume the required level of quality of functioning, as well as correct damaged programs and data. After failure, PS sometimes becomes inoperative for some time, the duration of which is determined by its recoverability. This requires computational resources and time to identify an inoperable state, diagnose the causes of the failure, as well as to implement recovery processes. The main indicators of the recovery process are its duration and probabilistic characteristics. Recoverability is also characterized by the complete restoration of the normal functioning of programs in the process of their manual or automatic restarting - restarting. The restart should ensure the resumption of the normal functioning of the PS, which requires computer resources and time, which can be characterized by a relative value (% of total resources).

Availability or availability: the property of the PS to be able to perform the desired function at a given point in time for a given

terms of use. Externally, accessibility can be assessed by the relative time during which the PS is in working condition, in proportion to the total time of use. Consequently, availability is a combination of completeness (on which the frequency of failures depends), error tolerance and recoverability, which together determine the duration of downtime for recovery after each failure, as well as the duration of time between failures. The generalization of the failure and recovery characteristics is made in the readiness coefficient criteria . This indicator reflects the probability of having recoverable programs and data in a healthy state at an arbitrary point in time.

The lower limit of the scale of reliability attributes in table 11.2 is reflected by the values ​​at which the functional suitability sharply decreases and the use of this type of PS becomes inconvenient, dangerous or unprofitable. An example of such worst, limit values ​​for many PS classes can be a mean time to failure of less than ten hours, a availability factor below 0.9, and a recovery time of more than ten minutes. On the other hand, the best values ​​of these attributes are practically limited to those resources that can be allocated to achieve them during development and operation. Computational and software resources of the object computer to directly ensure the reliability of the PS operation are usually in the range from 10% to 90%, with the latter values ​​corresponding to critical, especially highly reliable systems. Even for such critical software, the mean time between failures exceeds several thousand hours, the availability factor is no higher than 0.999, and the recovery time for failures is not less than a few seconds.

Efficiency: the ISO 9126 standard reflects two sub-characteristics of quality — temporal efficiency and use of computer resources, which are recommended to be described mainly by quantitative attributes characterizing the dynamics of the functioning of PS components. This standardized characteristic reflects only the private constructive efficiency of the use of computer resources, which should not be confused with the system efficiency of the functional fitness of PS when used in a particular system.

The main requirements for the characteristics of the characteristics of the temporal efficiency of the use of computing resources of the system are focused on the most critical indicators of p