Design and performance analysis of the Practical Fat Tree Network using a butterfly network

Youngsik Kim, Oh Young Kwon, Tack Don Han, Youngsong Mun

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The Practical Fat Tree Network (PFN) is proposed as an efficient interconnection structure for parallel processing computers using a butterfly network. It satisfies the concept of volume universality from Leiserson's theoretical fat tree network [1]. In the fat tree network, the locality of input traffic is a very important factor because it has significant effects on performance. The proposed performance model effectively highlights this locality factor. Performance is analyzed according to varing input traffic loads and buffer sizes. The accuracy of the model is confirmed by comparing its results with those from simulation. The proposed model is found to be very accurate. Although the PFN is a dynamic connection network, it shows low delay and high throughput for the applications that have locality.

Original languageEnglish
Pages (from-to)355-363
Number of pages9
JournalJournal of Systems Architecture
Volume43
Issue number1-5
Publication statusPublished - 1997 Jan 1

Fingerprint

Oils and fats
Throughput
Processing

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture

Cite this

Kim, Youngsik ; Kwon, Oh Young ; Han, Tack Don ; Mun, Youngsong. / Design and performance analysis of the Practical Fat Tree Network using a butterfly network. In: Journal of Systems Architecture. 1997 ; Vol. 43, No. 1-5. pp. 355-363.
@article{adf0389b6eac49768eecb362c49b0dbf,
title = "Design and performance analysis of the Practical Fat Tree Network using a butterfly network",
abstract = "The Practical Fat Tree Network (PFN) is proposed as an efficient interconnection structure for parallel processing computers using a butterfly network. It satisfies the concept of volume universality from Leiserson's theoretical fat tree network [1]. In the fat tree network, the locality of input traffic is a very important factor because it has significant effects on performance. The proposed performance model effectively highlights this locality factor. Performance is analyzed according to varing input traffic loads and buffer sizes. The accuracy of the model is confirmed by comparing its results with those from simulation. The proposed model is found to be very accurate. Although the PFN is a dynamic connection network, it shows low delay and high throughput for the applications that have locality.",
author = "Youngsik Kim and Kwon, {Oh Young} and Han, {Tack Don} and Youngsong Mun",
year = "1997",
month = "1",
day = "1",
language = "English",
volume = "43",
pages = "355--363",
journal = "Journal of Systems Architecture",
issn = "1383-7621",
publisher = "Elsevier",
number = "1-5",

}

Design and performance analysis of the Practical Fat Tree Network using a butterfly network. / Kim, Youngsik; Kwon, Oh Young; Han, Tack Don; Mun, Youngsong.

In: Journal of Systems Architecture, Vol. 43, No. 1-5, 01.01.1997, p. 355-363.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Design and performance analysis of the Practical Fat Tree Network using a butterfly network

AU - Kim, Youngsik

AU - Kwon, Oh Young

AU - Han, Tack Don

AU - Mun, Youngsong

PY - 1997/1/1

Y1 - 1997/1/1

N2 - The Practical Fat Tree Network (PFN) is proposed as an efficient interconnection structure for parallel processing computers using a butterfly network. It satisfies the concept of volume universality from Leiserson's theoretical fat tree network [1]. In the fat tree network, the locality of input traffic is a very important factor because it has significant effects on performance. The proposed performance model effectively highlights this locality factor. Performance is analyzed according to varing input traffic loads and buffer sizes. The accuracy of the model is confirmed by comparing its results with those from simulation. The proposed model is found to be very accurate. Although the PFN is a dynamic connection network, it shows low delay and high throughput for the applications that have locality.

AB - The Practical Fat Tree Network (PFN) is proposed as an efficient interconnection structure for parallel processing computers using a butterfly network. It satisfies the concept of volume universality from Leiserson's theoretical fat tree network [1]. In the fat tree network, the locality of input traffic is a very important factor because it has significant effects on performance. The proposed performance model effectively highlights this locality factor. Performance is analyzed according to varing input traffic loads and buffer sizes. The accuracy of the model is confirmed by comparing its results with those from simulation. The proposed model is found to be very accurate. Although the PFN is a dynamic connection network, it shows low delay and high throughput for the applications that have locality.

UR - http://www.scopus.com/inward/record.url?scp=0142213016&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0142213016&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0142213016

VL - 43

SP - 355

EP - 363

JO - Journal of Systems Architecture

JF - Journal of Systems Architecture

SN - 1383-7621

IS - 1-5

ER -