ES: a tool for predicting the performance of parallel systems

Author

Sinclair, J.B. ; Dawkins, W.P.

Author_Institution

Dept. of Electr. & Comput. Eng., Rice Univ., Houston, TX, USA

fYear

1994

fDate

31 Jan-2 Feb 1994

Firstpage

164

Lastpage

168

Abstract

ES is a tool for estimating the execution times of parallel algorithms on MIMD parallel systems. ES allows the user to model arbitrary task execution times, explicit task precedence and synchronization constraints, resource contention among tasks, and a variety of scheduling policies for shared resources. Given a model of a parallel algorithm and a parallel system, ES constructs a sequencing tree that represents some or all of the possible sequences of events that may occur during the execution of the algorithm on the system, and uses it to estimate the mean and standard deviation of the execution time of the parallel algorithm. The authors compare estimates generated by ES to measurements made of a parallel mergesort executing on an Intel iPSC/860 hypercube

Keywords

hypercube networks; parallel algorithms; parallel architectures; performance evaluation; scheduling; software tools; virtual machines; ES; Intel iPSC/860 hypercube; MIMD; execution time; model; parallel algorithm; parallel mergesort; parallel processing; parallel system; performance; resource contention; scheduling policy; sequencing tree; shared resource; software tool; synchronization constraint; task precedence; Algorithm design and analysis; Application software; Concurrent computing; Hypercubes; Parallel algorithms; Petri nets; Power generation; Power measurement; Processor scheduling; Steady-state;

fLanguage

English

Publisher

ieee

Conference_Titel

Modeling, Analysis, and Simulation of Computer and Telecommunication Systems, 1994., MASCOTS '94., Proceedings of the Second International Workshop on

Conference_Location

Durham, NC

Print_ISBN

0-8186-5292-6

Type

conf

DOI

10.1109/MASCOT.1994.284428

Filename

284428