Towards a communication characterization methodology for parallel applications

Sucheta Chodnekar, Viji Srinivasan, Aniruddha S. Vaidya, Anand Sivasubramaniam, Chitaranjan Das

Research output: Contribution to conferencePaper

28 Citations (Scopus)

Abstract

Interconnection network (ICN) is a vital component of a parallel machine and is often the limiting factor in the performance of several parallel applications. While ICN performance evaluation has been a widely researched topic, there have been very few studies that have used real applications to drive this research. In this paper, we develop a framework for characterizing the communication properties of parallel applications. Message generation frequency, spatial distribution of messages and message length are the three attributes that quantify any communication. We develop a methodology to quantify these attributes, in particular the first two attributes. We employ two strategies, namely dynamic and static, in our methodology. In the former, the applications are executed on an execution-driven simulator called SPASM, while in the latter they are executed on a parallel machine, IBM SP2. We gather communication events from these executions and feed them to a 2-D mesh network simulator. The log of the network activity is then analyzed using a statistical analysis package (SAS) to find the message inter-arrival time distribution and spatial distribution via regression analysis. Five shared memory applications and two message passing applications are analyzed to quantify their communication workloads. It is shown that it is possible to express the message generation and spatial distribution of an application in terms of commonly used distributions. These distributions can be used in the analysis of ICNs for developing realistic performance models.

Original languageEnglish (US)
Pages310-319
Number of pages10
StatePublished - Jan 1 1997
EventProceedings of the 1997 3rd International Symposium on High-Performance Computer Architecture, HPCA - San Antonio, TX, USA
Duration: Feb 1 1997Feb 5 1997

Other

OtherProceedings of the 1997 3rd International Symposium on High-Performance Computer Architecture, HPCA
CitySan Antonio, TX, USA
Period2/1/972/5/97

Fingerprint

Communication
Spatial distribution
Simulators
Message passing
Network performance
Regression analysis
Statistical methods
Data storage equipment

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Cite this

Chodnekar, S., Srinivasan, V., Vaidya, A. S., Sivasubramaniam, A., & Das, C. (1997). Towards a communication characterization methodology for parallel applications. 310-319. Paper presented at Proceedings of the 1997 3rd International Symposium on High-Performance Computer Architecture, HPCA, San Antonio, TX, USA, .
Chodnekar, Sucheta ; Srinivasan, Viji ; Vaidya, Aniruddha S. ; Sivasubramaniam, Anand ; Das, Chitaranjan. / Towards a communication characterization methodology for parallel applications. Paper presented at Proceedings of the 1997 3rd International Symposium on High-Performance Computer Architecture, HPCA, San Antonio, TX, USA, .10 p.
@conference{af7c01aba17f4673997cc4d1d750d54b,
title = "Towards a communication characterization methodology for parallel applications",
abstract = "Interconnection network (ICN) is a vital component of a parallel machine and is often the limiting factor in the performance of several parallel applications. While ICN performance evaluation has been a widely researched topic, there have been very few studies that have used real applications to drive this research. In this paper, we develop a framework for characterizing the communication properties of parallel applications. Message generation frequency, spatial distribution of messages and message length are the three attributes that quantify any communication. We develop a methodology to quantify these attributes, in particular the first two attributes. We employ two strategies, namely dynamic and static, in our methodology. In the former, the applications are executed on an execution-driven simulator called SPASM, while in the latter they are executed on a parallel machine, IBM SP2. We gather communication events from these executions and feed them to a 2-D mesh network simulator. The log of the network activity is then analyzed using a statistical analysis package (SAS) to find the message inter-arrival time distribution and spatial distribution via regression analysis. Five shared memory applications and two message passing applications are analyzed to quantify their communication workloads. It is shown that it is possible to express the message generation and spatial distribution of an application in terms of commonly used distributions. These distributions can be used in the analysis of ICNs for developing realistic performance models.",
author = "Sucheta Chodnekar and Viji Srinivasan and Vaidya, {Aniruddha S.} and Anand Sivasubramaniam and Chitaranjan Das",
year = "1997",
month = "1",
day = "1",
language = "English (US)",
pages = "310--319",
note = "Proceedings of the 1997 3rd International Symposium on High-Performance Computer Architecture, HPCA ; Conference date: 01-02-1997 Through 05-02-1997",

}

Chodnekar, S, Srinivasan, V, Vaidya, AS, Sivasubramaniam, A & Das, C 1997, 'Towards a communication characterization methodology for parallel applications' Paper presented at Proceedings of the 1997 3rd International Symposium on High-Performance Computer Architecture, HPCA, San Antonio, TX, USA, 2/1/97 - 2/5/97, pp. 310-319.

Towards a communication characterization methodology for parallel applications. / Chodnekar, Sucheta; Srinivasan, Viji; Vaidya, Aniruddha S.; Sivasubramaniam, Anand; Das, Chitaranjan.

1997. 310-319 Paper presented at Proceedings of the 1997 3rd International Symposium on High-Performance Computer Architecture, HPCA, San Antonio, TX, USA, .

Research output: Contribution to conferencePaper

TY - CONF

T1 - Towards a communication characterization methodology for parallel applications

AU - Chodnekar, Sucheta

AU - Srinivasan, Viji

AU - Vaidya, Aniruddha S.

AU - Sivasubramaniam, Anand

AU - Das, Chitaranjan

PY - 1997/1/1

Y1 - 1997/1/1

N2 - Interconnection network (ICN) is a vital component of a parallel machine and is often the limiting factor in the performance of several parallel applications. While ICN performance evaluation has been a widely researched topic, there have been very few studies that have used real applications to drive this research. In this paper, we develop a framework for characterizing the communication properties of parallel applications. Message generation frequency, spatial distribution of messages and message length are the three attributes that quantify any communication. We develop a methodology to quantify these attributes, in particular the first two attributes. We employ two strategies, namely dynamic and static, in our methodology. In the former, the applications are executed on an execution-driven simulator called SPASM, while in the latter they are executed on a parallel machine, IBM SP2. We gather communication events from these executions and feed them to a 2-D mesh network simulator. The log of the network activity is then analyzed using a statistical analysis package (SAS) to find the message inter-arrival time distribution and spatial distribution via regression analysis. Five shared memory applications and two message passing applications are analyzed to quantify their communication workloads. It is shown that it is possible to express the message generation and spatial distribution of an application in terms of commonly used distributions. These distributions can be used in the analysis of ICNs for developing realistic performance models.

AB - Interconnection network (ICN) is a vital component of a parallel machine and is often the limiting factor in the performance of several parallel applications. While ICN performance evaluation has been a widely researched topic, there have been very few studies that have used real applications to drive this research. In this paper, we develop a framework for characterizing the communication properties of parallel applications. Message generation frequency, spatial distribution of messages and message length are the three attributes that quantify any communication. We develop a methodology to quantify these attributes, in particular the first two attributes. We employ two strategies, namely dynamic and static, in our methodology. In the former, the applications are executed on an execution-driven simulator called SPASM, while in the latter they are executed on a parallel machine, IBM SP2. We gather communication events from these executions and feed them to a 2-D mesh network simulator. The log of the network activity is then analyzed using a statistical analysis package (SAS) to find the message inter-arrival time distribution and spatial distribution via regression analysis. Five shared memory applications and two message passing applications are analyzed to quantify their communication workloads. It is shown that it is possible to express the message generation and spatial distribution of an application in terms of commonly used distributions. These distributions can be used in the analysis of ICNs for developing realistic performance models.

UR - http://www.scopus.com/inward/record.url?scp=0030826853&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030826853&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0030826853

SP - 310

EP - 319

ER -

Chodnekar S, Srinivasan V, Vaidya AS, Sivasubramaniam A, Das C. Towards a communication characterization methodology for parallel applications. 1997. Paper presented at Proceedings of the 1997 3rd International Symposium on High-Performance Computer Architecture, HPCA, San Antonio, TX, USA, .