Algorithmic Control in Concurrent Computations

Abstract: In this paper, functioning and interaction of distributed devices and concurrent algorithms are analyzed in the context of the theory of algorithms. Although different systems in a network can function independently, their interaction is usually subjected to definite rules. These rules often form algorithms of interaction for distributed systems and devices. The main goal is to understand how control algorithms in distributed systems organize computational processes. Characteristics of distributed computational processes are considered from the algorithmic perspective. Conditions are found when such a network controlled by algorithm functions as a single algorithm and types of such unifying algorithms are determined. All mathematical models of distributed computation, including such popular structures as neural networks, Petri nets, systolic arrays, iterative arrays, and cellular automata, are synthesized in the concept of grid automaton. These automata are used as a basic tool for studying distributed devices and concurrent algorithms