Toward Incremental Parallelization Using Navigational Programming

Lei PAN; Wenhui ZHANG; Arthur ASUNCION; Ming Kin LAI; Michael B. DILLENCOURT; Lubomir F. BIC; Laurence T. YANG

doi:10.1093/ietisy/e89-d.2.390

Toward Incremental Parallelization Using Navigational Programming

Lei PAN, Wenhui ZHANG, Arthur ASUNCION, Ming Kin LAI, Michael B. DILLENCOURT, Lubomir F. BIC, Laurence T. YANG

Full Text Views

0

Share
Cite this

Summary :

The Navigational Programming (NavP) methodology is based on the principle of self-migrating computations. It is a truly incremental methodology for developing parallel programs: each step represents a functioning program, and each intermediate program is an improvement over its predecessor. The transformations are mechanical and straightforward to apply. We illustrate our methodology in the context of matrix multiplication, showing how the transformations lead from a sequential program to a fully parallel program. The NavP methodology is conducive to new ways of thinking that lead to ease of programming and high performance. Even though our parallel algorithm was derived using a sequence of mechanical transformations, it displays certain performance advantages over the classical handcrafted Gentleman's Algorithm.

Publication: IEICE TRANSACTIONS on Information Vol.E89-D No.2 pp.390-398

Publication Date: 2006/02/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1093/ietisy/e89-d.2.390

Type of Manuscript: Special Section PAPER (Special Section on Parallel/Distributed Computing and Networking)

Category: Parallel/Distributed Programming Models, Paradigms and Tools

Cite this

Copy

Lei PAN, Wenhui ZHANG, Arthur ASUNCION, Ming Kin LAI, Michael B. DILLENCOURT, Lubomir F. BIC, Laurence T. YANG, "Toward Incremental Parallelization Using Navigational Programming" in IEICE TRANSACTIONS on Information, vol. E89-D, no. 2, pp. 390-398, February 2006, doi: 10.1093/ietisy/e89-d.2.390.
Abstract: The Navigational Programming (NavP) methodology is based on the principle of self-migrating computations. It is a truly incremental methodology for developing parallel programs: each step represents a functioning program, and each intermediate program is an improvement over its predecessor. The transformations are mechanical and straightforward to apply. We illustrate our methodology in the context of matrix multiplication, showing how the transformations lead from a sequential program to a fully parallel program. The NavP methodology is conducive to new ways of thinking that lead to ease of programming and high performance. Even though our parallel algorithm was derived using a sequence of mechanical transformations, it displays certain performance advantages over the classical handcrafted Gentleman's Algorithm.
URL: https://globals.ieice.org/en_transactions/information/10.1093/ietisy/e89-d.2.390/_p

Copy

@ARTICLE{e89-d_2_390,
author={Lei PAN, Wenhui ZHANG, Arthur ASUNCION, Ming Kin LAI, Michael B. DILLENCOURT, Lubomir F. BIC, Laurence T. YANG, },
journal={IEICE TRANSACTIONS on Information},
title={Toward Incremental Parallelization Using Navigational Programming},
year={2006},
volume={E89-D},
number={2},
pages={390-398},
abstract={The Navigational Programming (NavP) methodology is based on the principle of self-migrating computations. It is a truly incremental methodology for developing parallel programs: each step represents a functioning program, and each intermediate program is an improvement over its predecessor. The transformations are mechanical and straightforward to apply. We illustrate our methodology in the context of matrix multiplication, showing how the transformations lead from a sequential program to a fully parallel program. The NavP methodology is conducive to new ways of thinking that lead to ease of programming and high performance. Even though our parallel algorithm was derived using a sequence of mechanical transformations, it displays certain performance advantages over the classical handcrafted Gentleman's Algorithm.},
keywords={},
doi={10.1093/ietisy/e89-d.2.390},
ISSN={1745-1361},
month={February},}

Copy

TY - JOUR
TI - Toward Incremental Parallelization Using Navigational Programming
T2 - IEICE TRANSACTIONS on Information
SP - 390
EP - 398
AU - Lei PAN
AU - Wenhui ZHANG
AU - Arthur ASUNCION
AU - Ming Kin LAI
AU - Michael B. DILLENCOURT
AU - Lubomir F. BIC
AU - Laurence T. YANG
PY - 2006
DO - 10.1093/ietisy/e89-d.2.390
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E89-D
IS - 2
JA - IEICE TRANSACTIONS on Information
Y1 - February 2006
AB - The Navigational Programming (NavP) methodology is based on the principle of self-migrating computations. It is a truly incremental methodology for developing parallel programs: each step represents a functioning program, and each intermediate program is an improvement over its predecessor. The transformations are mechanical and straightforward to apply. We illustrate our methodology in the context of matrix multiplication, showing how the transformations lead from a sequential program to a fully parallel program. The NavP methodology is conducive to new ways of thinking that lead to ease of programming and high performance. Even though our parallel algorithm was derived using a sequence of mechanical transformations, it displays certain performance advantages over the classical handcrafted Gentleman's Algorithm.
ER -