РІВНОМІРНИЙ РОЗПОДІЛ РЕСУРСІВ КОМП'ЮТЕРНИХ СИСТЕМ, ЩО МАЮТЬ ГІПЕРКОНВЕРГЕНТНУ ІНФРАСТРУКТУРУ

Citations Over Time

Abstract

The relevance of research. The infrastructure created on a converged platform involves the integration of memory, computing and network resources into a single pool, and with a hyper-convergent infrastructure, computing power, storage, servers, networks are combined into one whole using software tools. The purpose of the article is to develop an approach to reducing the uneven distribution of resources in computer networks of hyperconvergent architecture. The results of the study. A method is proposed for minimizing the average transaction latency in computer networks of a hyperconvergent architecture, which makes it possible to construct a uniform distribution of the allocated computational resources for processing a set of system tasks using the quanta of a given time interval. The objective function of the task of finding a rational partition of the set of queries processed in the computer network into subsets and their distribution among the nodes of the hyperconvergent network is determined by entering the penalty function when each query unit is allocated a computing resource in the current time slot. Conclusions. An approach to the formation of a uniform distribution of resources in computer networks of hyperconvergent architecture is proposed. The application of the approach allows you to balance the network load with a large number of requests and to achieve the requirements for the efficiency of their processing.

Related Papers

• → XML and MPEG-7 for Interactive Annotation and Retrieval using Semantic Meta-data.(2002)8 cited
• → A Dataflow Graph Transformation Language and Query Rewriting System for RDF Ontologies(2012)3 cited
• → A hierarchical parallelizing compiler for VLIW/MIMD machines(1993)3 cited
• → An Integrated Runtime Scheduler for MPI(2012)
• → Techniques for the efficient analysis of cache performance(1996)