1991Physica ScriptaOpen access

Computation on Parallel Message-Passing Computers

J. Petersen

Open full text 0 citations

Abstract

Within parallel computing, one may divide the present architectures into 3 categories, (a) massively parallel SIMD, (b) Shared memory MIMD and (c) Distributed memory MIMD (or Message Passing). (SIMD = Single Instruction Multiple Data; MIMD = Multiple Instruction Multiple Data). The "simplest" of these is the message-passing architecture. A Message-Passing system consists of several processors, where each processor has a "fair" amount of local memory and is connected to some kind of communication network. This talk covers some of the most common tools of an established parallel message passing operating system, Trollius [1]. One tool, TGraph [2], developed at Bergen Scientific Centre, is a tool for graphics output under X-windows from the processors. A programming philosophy [3] for message-passing computers which is based on `domain decomposition', and some specific applications, such as Wave Equation [4], Gauss Elimination [5] and a large Climate Model project are discussed.

Open-access reader

About this research paper

What this paper is about

Within parallel computing, one may divide the present architectures into 3 categories, (a) massively parallel SIMD, (b) Shared memory MIMD and (c) Distributed memory MIMD (or Message Passing). (SIMD = Single Instruction Multiple Data; MIMD = Multiple Instruction Multiple Data). The "simplest" of these is the message-passing architecture. A Message-Passing system consists of several processors, where each processor has a "fair" amount of local memory and is connected to some kind of communication network. This talk covers some of the most common tools of an established parallel message passing operating system, Trollius [1]. One tool, TGraph [2], developed at Bergen Scientific Centre, is a tool for graphics output under X-windows from the processors. A programming philosophy [3] for message-passing computers which is based on `domain decomposition', and some specific applications, such as Wave Equation [4], Gauss Elimination [5] and a large Climate Model project are discussed.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Within parallel computing, one may divide the present architectures into 3 categories, (a) massively parallel SIMD, (b) Shared memory MIMD and (c) Distributed memory MIMD (or Message Passing). (SIMD = Single Instruction Multiple Data; MIMD = Multiple Instruction Multiple Data). The "simplest" of these is the message-passing architecture. A Message-Passing system consists of several processors, where each processor has a "fair" amount of local memory and is connected to some kind of communication network. This talk covers some of the most common tools of an established parallel message passing operating system, Trollius [1]. One tool, TGraph [2], developed at Bergen Scientific Centre, is a tool for graphics output under X-windows from the processors. A programming philosophy [3] for message-passing computers which is based on `domain decomposition', and some specific applications, such as Wave Equation [4], Gauss Elimination [5] and a large Climate Model project are discussed.

Key concepts: MIMD, Computer science, Parallel computing, SIMD, Massively parallel, Computation, Message passing, Distributed memory

Related papers

Back to paper searchBrowse research topicsOriginal source
Computation on Parallel Message-Passing Computers — Research Paper | ScholarLens