Fault-Tolerant Operation of an Integer Data Processing System

The article proposes for consideration the process of functioning of a computer system (CS) in a system of residual classes (SRC) based on the use of the principle of fault tolerance. The fault tolerance of a computer system is a certain characteristic of the CS architecture, ensuring it with the ability to operate uninterruptedly in accordance with the executable program. The fault-tolerance property provides the ability to perform the given computing features after the emergence of failures by reducing any performance quality indicators within acceptable limits and without deteriorating the performance quality indicators of the CS. The structure and properties of the functioning of a fault-tolerant CS in the SRC are considered. Variants of algorithms for functioning of fault-tolerant CSs in the SRC in replacement and gradual degradation modes are presented, as well as a graph of CS functioning in the SRC in replacement mode with subsequent gradual degradation is presented. In order to formalize the study of the process of functioning of the CS in the SRC, the article introduces an (n + 1)-bit binary code that displays a set of signs of a healthy state of the computing tracts of the CS. By means of this code, an analysis of various modes of fault-tolerant operation of the CS in the SRC was made.