View in EDS

Design of Dynamic Network for Parallel Processing on a Distributed System.

Saved in:
Bibliographic Details
Title: Design of Dynamic Network for Parallel Processing on a Distributed System.
Authors: Ebrahem, Aseel Thamer, Mahmmood, Basil Shukr
Source: Ingénierie des Systèmes d'Information; Feb2023, Vol. 28 Issue 1, p239-245, 7p
Subject Terms: PARALLEL processing, COMPUTER software
Abstract: Modern computation systems involve multicomputer configurations. Multiple computers enable multiple threads to be executed simultaneously, with the ability to perform the same operations on different processors (computers) at the same time. This paper addresses the building of a software application to be implemented on 8*8 dynamic multistage network exchange depending on the client/server principles so that one can select one type of different topologies that is suitable to the type of application. The network can serve any number of nodes. Two different applications were examined, convolution operations and matrix multiplication. The goal of this paper is to explore the different ways, in which the multistage network topology can simulate supercomputer systems employing large-scale parallel processing. This paper proposes designing parallel systems on a distributed system, running different topologies such as linear systolic, mesh, and hypercube topologies of the parallel processor's networks, and also a dynamic selection of the appropriate network topology based on the nature of the solved problem. Simulation of parallel processing systems on distributed environments mainly done through Socket programming based on JAVA threads. [ABSTRACT FROM AUTHOR]
Copyright of Ingénierie des Systèmes d'Information is the property of International Information & Engineering Technology Association (IIETA) and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Complementary Index
Description
Abstract:Modern computation systems involve multicomputer configurations. Multiple computers enable multiple threads to be executed simultaneously, with the ability to perform the same operations on different processors (computers) at the same time. This paper addresses the building of a software application to be implemented on 8*8 dynamic multistage network exchange depending on the client/server principles so that one can select one type of different topologies that is suitable to the type of application. The network can serve any number of nodes. Two different applications were examined, convolution operations and matrix multiplication. The goal of this paper is to explore the different ways, in which the multistage network topology can simulate supercomputer systems employing large-scale parallel processing. This paper proposes designing parallel systems on a distributed system, running different topologies such as linear systolic, mesh, and hypercube topologies of the parallel processor's networks, and also a dynamic selection of the appropriate network topology based on the nature of the solved problem. Simulation of parallel processing systems on distributed environments mainly done through Socket programming based on JAVA threads. [ABSTRACT FROM AUTHOR]
ISSN:16331311
DOI:10.18280/isi.280128