Abstract
The article is devoted to the problem of parallel algorithms and their practical implementations. On the basis of parallel computer analysis in the world the parallel systems are divided into two basic groups - synchronous and asynchronous systems - which are very different from the system point of view. This article describes the development of real parallel algorithms for Jacobi iteration. This individual practical example demonstrates the influence of decomposition strategies for performance evaluation of parallel Jacobi iteration and discusses the ways for their parallel implementations.
Highlights
For the contemporary technical and programmable level of the reachable computer means the use of various forms of basic principles of the parallel activity is dominant [8, 15, 16]
The second level creates the user developing programming environments, which support the development of the modular application programs as the basic condition to their potential parallel activity
This parallel support goes in this time up to the level of the elementary program elements in the form of the objects (OOP – object oriented programming)
Summary
For the contemporary technical and programmable level of the reachable computer means (personal computers, minicomputers, supercomputers, etc.) the use of various forms of basic principles of the parallel activity is dominant [8, 15, 16]. The second level creates the user developing programming environments, which support the development of the modular application programs as the basic condition to their potential parallel activity. The basic system properties are given through the existence of some kind of the common shared memory M by parallel processors Pi, which in substantial measure simplifies their application programming using. This group covers the field of various forms of computer networks. When designing a parallel program the description of the highlevel algorithm must include the method we intend to use to divide the application into processes and to distribute data to different nodes – the decomposition strategy. To the performance evaluation we briefly review the techniques most commonly adopted for the evaluation of parallel systems and its metrics
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: Communications - Scientific letters of the University of Zilina
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.