Annotation: The article proposes a model and method for optimizing computing processes in computing systems with a parallel architecture. The construction of the model consists in dividing computer programs for solving problems into relatively independent elements (blocks) and representing the information dependence of the blocks in the form of corresponding oriented graphs. The organization of the computational process in this case consists in the dynamic distribution of the resources of the computing system for implementing the operations of each of the blocks. As an indicator of the effectiveness of this process, the time taken to implement the operations of all blocks of the programs under consideration (the time to solve the corresponding tasks) was taken. The goal of optimization is to shorten the time as much as possible. To form an optimal resource allocation, a method based on the branch and bound procedure is proposed.

Keywords: computer, computing system with parallel architecture, organization of computing process, optimization, model, method