Detail publikace
Predicting performance of SMP clusters
DVOŘÁK, V., STAROBA, J., ČEJKA, R.
Originální název
Predicting performance of SMP clusters
Anglický název
Predicting performance of SMP clusters
Jazyk
en
Originální abstrakt
The paper reports on simulation-based performance prediction of parallel applications running on SMP clusters. First the typical message-passing programs are investigated, which demonstrate different computation/ communication patterns: process farm, pipeline and data parallel computing on a linear processor chain. The usefulness of various SMP cluster configurations is discussed with regard to obtained results. Then the message-passing model of shared memory and SMP clusters is used to simulate execution of hybrid message-passing/shared objects programs. A large system of linear equations is used as an example. A single SMP, a SMP cluster and a workstation cluster are compared in performance with a limited L1-cache size.
Anglický abstrakt
The paper reports on simulation-based performance prediction of parallel applications running on SMP clusters. First the typical message-passing programs are investigated, which demonstrate different computation/ communication patterns: process farm, pipeline and data parallel computing on a linear processor chain. The usefulness of various SMP cluster configurations is discussed with regard to obtained results. Then the message-passing model of shared memory and SMP clusters is used to simulate execution of hybrid message-passing/shared objects programs. A large system of linear equations is used as an example. A single SMP, a SMP cluster and a workstation cluster are compared in performance with a limited L1-cache size.
Dokumenty
BibTex
@inbook{BUT54888,
author="Rudolf {Čejka} and Václav {Dvořák} and Jiří {Staroba}",
title="Predicting performance of SMP clusters",
annote="The paper reports on simulation-based performance prediction of parallel applications running on SMP clusters. First the typical message-passing programs are investigated, which demonstrate different computation/ communication patterns: process farm, pipeline and data parallel computing on a linear processor chain. The usefulness of various SMP cluster configurations is discussed with regard to obtained results. Then the message-passing model of shared memory and SMP clusters is used to simulate execution of hybrid message-passing/shared objects programs. A large system of linear equations is used as an example. A single SMP, a SMP cluster and a workstation cluster are compared in performance with a limited L1-cache size.",
address="Kluwer Academic Publishers",
booktitle="Distributed and Parallel Systems - Cluster and Grid Computing",
chapter="54888",
institution="Kluwer Academic Publishers",
year="2002",
month="september",
pages="38--45",
publisher="Kluwer Academic Publishers",
type="book chapter"
}