BSP vs LogP

Gianfranco Bilardi; Kieran T. Herley; Andrea Pietracaprina; Geppino Pucci; Paul Spirakis

doi:10.1145/237502.237504

BSP vs LogP

Gianfranco Bilardi
, Kieran T. Herley
, Andrea Pietracaprina
, Geppino Pucci
, Paul Spirakis

University of Padua

Research output: Contribution to conference › Paper › peer-review

Abstract

A quantitative comparison of the BSP and LogP models for parallel computation is developed. Very efficient cross simulations between the two models are derived, showing their substantial equivalence for algorithmic design guided by asymptotic analysis. It is also shown that the two models can be implemented with similar performance on most point-to-point networks. In conclusion, within the limits of our analysis that is mainly of asymptotic nature, BSP and LogP can be viewed as closely related variants within the bandwidth-latency framework for modeling parallel computation. BSP seems somewhat preferable due to greater simplicity and portability, and slightly greater power.

Original language	English
Pages	25-32
Number of pages	8
DOIs	https://doi.org/10.1145/237502.237504
Publication status	Published - 1996
Externally published	Yes
Event	Proceedings of the 1996 8th Annual ACM Symposium on Parallel Algorithms and Architectures - Padua, Italy Duration: 24 Jun 1996 → 26 Jun 1996

Conference

Conference	Proceedings of the 1996 8th Annual ACM Symposium on Parallel Algorithms and Architectures
City	Padua, Italy
Period	24/06/96 → 26/06/96

Access to Document

10.1145/237502.237504

Cite this

@conference{c35e23ae7021423eace294d96251f05b,

title = "BSP vs LogP",

abstract = "A quantitative comparison of the BSP and LogP models for parallel computation is developed. Very efficient cross simulations between the two models are derived, showing their substantial equivalence for algorithmic design guided by asymptotic analysis. It is also shown that the two models can be implemented with similar performance on most point-to-point networks. In conclusion, within the limits of our analysis that is mainly of asymptotic nature, BSP and LogP can be viewed as closely related variants within the bandwidth-latency framework for modeling parallel computation. BSP seems somewhat preferable due to greater simplicity and portability, and slightly greater power.",

author = "Gianfranco Bilardi and Herley, \{Kieran T.\} and Andrea Pietracaprina and Geppino Pucci and Paul Spirakis",

year = "1996",

doi = "10.1145/237502.237504",

language = "English",

pages = "25--32",

note = "Proceedings of the 1996 8th Annual ACM Symposium on Parallel Algorithms and Architectures ; Conference date: 24-06-1996 Through 26-06-1996",

}

TY - CONF

T1 - BSP vs LogP

AU - Bilardi, Gianfranco

AU - Herley, Kieran T.

AU - Pietracaprina, Andrea

AU - Pucci, Geppino

AU - Spirakis, Paul

PY - 1996

Y1 - 1996

N2 - A quantitative comparison of the BSP and LogP models for parallel computation is developed. Very efficient cross simulations between the two models are derived, showing their substantial equivalence for algorithmic design guided by asymptotic analysis. It is also shown that the two models can be implemented with similar performance on most point-to-point networks. In conclusion, within the limits of our analysis that is mainly of asymptotic nature, BSP and LogP can be viewed as closely related variants within the bandwidth-latency framework for modeling parallel computation. BSP seems somewhat preferable due to greater simplicity and portability, and slightly greater power.

AB - A quantitative comparison of the BSP and LogP models for parallel computation is developed. Very efficient cross simulations between the two models are derived, showing their substantial equivalence for algorithmic design guided by asymptotic analysis. It is also shown that the two models can be implemented with similar performance on most point-to-point networks. In conclusion, within the limits of our analysis that is mainly of asymptotic nature, BSP and LogP can be viewed as closely related variants within the bandwidth-latency framework for modeling parallel computation. BSP seems somewhat preferable due to greater simplicity and portability, and slightly greater power.

UR - https://www.scopus.com/pages/publications/0030416441

U2 - 10.1145/237502.237504

DO - 10.1145/237502.237504

M3 - Paper

AN - SCOPUS:0030416441

SP - 25

EP - 32

T2 - Proceedings of the 1996 8th Annual ACM Symposium on Parallel Algorithms and Architectures

Y2 - 24 June 1996 through 26 June 1996

ER -

BSP vs LogP

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this