Algorithm Transformation Techniques for Concurrent Processors

Research output: Contribution to journalArticle

112 Citations (Scopus)

Abstract

Progress in supercomputer technology leads to two major trends. First, many of the existing algorithms will need to be redesigned for efficient concurrent implementation using supercomputers. Second, a continuous increase will be apparent in the number of application-specific VLSI integrated circuits, which can provide the performance of supercomputers, using single chips or chipsets (at the expense of design time for algorithm and architecture development). Both of these approaches require considerable efforts in the development of algorithms for specific applications. This paper reviews four independent algorithm transformation methodologies: program unfolding, retiming, lookahead algorithms, and index mapping transformations. These transformation techniques exploit the available parallelism in iterative dataflow programs and create additional parallelism if necessary.

Original languageEnglish (US)
Pages (from-to)1879-1895
Number of pages17
JournalProceedings of the IEEE
Volume77
Issue number12
DOIs
StatePublished - Jan 1 1989

Fingerprint

Supercomputers
VLSI circuits
Application specific integrated circuits

Cite this

Algorithm Transformation Techniques for Concurrent Processors. / Parhi, Keshab K.

In: Proceedings of the IEEE, Vol. 77, No. 12, 01.01.1989, p. 1879-1895.

Research output: Contribution to journalArticle

@article{256471fa3ae8435594f795e3c30fa9bb,
title = "Algorithm Transformation Techniques for Concurrent Processors",
abstract = "Progress in supercomputer technology leads to two major trends. First, many of the existing algorithms will need to be redesigned for efficient concurrent implementation using supercomputers. Second, a continuous increase will be apparent in the number of application-specific VLSI integrated circuits, which can provide the performance of supercomputers, using single chips or chipsets (at the expense of design time for algorithm and architecture development). Both of these approaches require considerable efforts in the development of algorithms for specific applications. This paper reviews four independent algorithm transformation methodologies: program unfolding, retiming, lookahead algorithms, and index mapping transformations. These transformation techniques exploit the available parallelism in iterative dataflow programs and create additional parallelism if necessary.",
author = "Parhi, {Keshab K}",
year = "1989",
month = "1",
day = "1",
doi = "10.1109/5.48830",
language = "English (US)",
volume = "77",
pages = "1879--1895",
journal = "Proceedings of the IEEE",
issn = "0018-9219",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "12",

}

TY - JOUR

T1 - Algorithm Transformation Techniques for Concurrent Processors

AU - Parhi, Keshab K

PY - 1989/1/1

Y1 - 1989/1/1

N2 - Progress in supercomputer technology leads to two major trends. First, many of the existing algorithms will need to be redesigned for efficient concurrent implementation using supercomputers. Second, a continuous increase will be apparent in the number of application-specific VLSI integrated circuits, which can provide the performance of supercomputers, using single chips or chipsets (at the expense of design time for algorithm and architecture development). Both of these approaches require considerable efforts in the development of algorithms for specific applications. This paper reviews four independent algorithm transformation methodologies: program unfolding, retiming, lookahead algorithms, and index mapping transformations. These transformation techniques exploit the available parallelism in iterative dataflow programs and create additional parallelism if necessary.

AB - Progress in supercomputer technology leads to two major trends. First, many of the existing algorithms will need to be redesigned for efficient concurrent implementation using supercomputers. Second, a continuous increase will be apparent in the number of application-specific VLSI integrated circuits, which can provide the performance of supercomputers, using single chips or chipsets (at the expense of design time for algorithm and architecture development). Both of these approaches require considerable efforts in the development of algorithms for specific applications. This paper reviews four independent algorithm transformation methodologies: program unfolding, retiming, lookahead algorithms, and index mapping transformations. These transformation techniques exploit the available parallelism in iterative dataflow programs and create additional parallelism if necessary.

UR - http://www.scopus.com/inward/record.url?scp=0024883413&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024883413&partnerID=8YFLogxK

U2 - 10.1109/5.48830

DO - 10.1109/5.48830

M3 - Article

VL - 77

SP - 1879

EP - 1895

JO - Proceedings of the IEEE

JF - Proceedings of the IEEE

SN - 0018-9219

IS - 12

ER -