Architectural differences of efficient sequential and parallel computers

Research output: Contribution to journalArticleScientificpeer-review

9 Citations (Scopus)

Abstract

In this paper we try to conclude what kind of a computer architecture is efficient for executing sequential problems, and what kind of an architecture is efficient for executing parallel problems from the processor architect's point of view. For that purpose we analytically evaluate the performance of eight general purpose processor architectures representing widely both commercial and scientific processor designs in both single processor and multiprocessor setups. The results are interesting. The most efficient architecture for sequential problems is a two-level pipelined VLIW architecture with few parallel functional units. The most efficient architecture for parallel problems is a deeply inter-thread superpipelined architecture in which functional units are chained. Thus, designing a computer for efficient sequential computation leads to a very different architecture than designing one for efficient parallel computation and there exists no single optimal architecture for general purpose computation.
Original languageEnglish
Pages (from-to)1017-1041
JournalJournal of Systems Architecture
Volume47
Issue number13
DOIs
Publication statusPublished - 2002
MoE publication typeA1 Journal article-refereed

Fingerprint

Very long instruction word architecture
Computer architecture

Keywords

  • Computer architecture
  • Sequential computation
  • Parallel computation
  • Processor architecture
  • Pipeline organization

Cite this

@article{b0e3797f78874232b4aa7ef852d4d8ed,
title = "Architectural differences of efficient sequential and parallel computers",
abstract = "In this paper we try to conclude what kind of a computer architecture is efficient for executing sequential problems, and what kind of an architecture is efficient for executing parallel problems from the processor architect's point of view. For that purpose we analytically evaluate the performance of eight general purpose processor architectures representing widely both commercial and scientific processor designs in both single processor and multiprocessor setups. The results are interesting. The most efficient architecture for sequential problems is a two-level pipelined VLIW architecture with few parallel functional units. The most efficient architecture for parallel problems is a deeply inter-thread superpipelined architecture in which functional units are chained. Thus, designing a computer for efficient sequential computation leads to a very different architecture than designing one for efficient parallel computation and there exists no single optimal architecture for general purpose computation.",
keywords = "Computer architecture, Sequential computation, Parallel computation, Processor architecture, Pipeline organization",
author = "Martti Forsell",
year = "2002",
doi = "10.1016/S1383-7621(02)00064-4",
language = "English",
volume = "47",
pages = "1017--1041",
journal = "Journal of Systems Architecture",
issn = "1383-7621",
publisher = "Elsevier",
number = "13",

}

Architectural differences of efficient sequential and parallel computers. / Forsell, Martti.

In: Journal of Systems Architecture, Vol. 47, No. 13, 2002, p. 1017-1041.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Architectural differences of efficient sequential and parallel computers

AU - Forsell, Martti

PY - 2002

Y1 - 2002

N2 - In this paper we try to conclude what kind of a computer architecture is efficient for executing sequential problems, and what kind of an architecture is efficient for executing parallel problems from the processor architect's point of view. For that purpose we analytically evaluate the performance of eight general purpose processor architectures representing widely both commercial and scientific processor designs in both single processor and multiprocessor setups. The results are interesting. The most efficient architecture for sequential problems is a two-level pipelined VLIW architecture with few parallel functional units. The most efficient architecture for parallel problems is a deeply inter-thread superpipelined architecture in which functional units are chained. Thus, designing a computer for efficient sequential computation leads to a very different architecture than designing one for efficient parallel computation and there exists no single optimal architecture for general purpose computation.

AB - In this paper we try to conclude what kind of a computer architecture is efficient for executing sequential problems, and what kind of an architecture is efficient for executing parallel problems from the processor architect's point of view. For that purpose we analytically evaluate the performance of eight general purpose processor architectures representing widely both commercial and scientific processor designs in both single processor and multiprocessor setups. The results are interesting. The most efficient architecture for sequential problems is a two-level pipelined VLIW architecture with few parallel functional units. The most efficient architecture for parallel problems is a deeply inter-thread superpipelined architecture in which functional units are chained. Thus, designing a computer for efficient sequential computation leads to a very different architecture than designing one for efficient parallel computation and there exists no single optimal architecture for general purpose computation.

KW - Computer architecture

KW - Sequential computation

KW - Parallel computation

KW - Processor architecture

KW - Pipeline organization

U2 - 10.1016/S1383-7621(02)00064-4

DO - 10.1016/S1383-7621(02)00064-4

M3 - Article

VL - 47

SP - 1017

EP - 1041

JO - Journal of Systems Architecture

JF - Journal of Systems Architecture

SN - 1383-7621

IS - 13

ER -