A study on snoop cache systems for single-chip multiprocessors

Takuya Terasawa, Keisuke Inoue, Hitoshi Kurosawa, Hideharu Amano

Research output: Contribution to journalArticle

Abstract

Progress in device design and implementation technologies will change the structure of bus-connected multiprocessors and allow them to be implemented on a single chip. In such an implementation, the speed of the bus inside the chip is far faster than that of the backplane bus, and data transfer between the chip and external devices will become a bottleneck. Many studies on snoop cache protocols have been made, but they generally assume that the cache memory is on a printed circuit board. In this paper, we first classify the snoop cache protocols in terms of accesses to off-chip shared memory, which will be the principal cause of performance degradation, and then compare them quantitatively. Evaluations are made with an instruction-level multiprocessor simulator and practical parallel applications, varying the cache size or the access latency of shared memory. The results show that an I/N/C protocol that actively uses the line transfers between caches achieves the highest performance under all conditions.

Original languageEnglish
Pages (from-to)62-72
Number of pages11
JournalSystems and Computers in Japan
Volume28
Issue number2
Publication statusPublished - 1998 Feb

Fingerprint

Chip multiprocessors
Cache
Chip
Shared Memory
Multiprocessor
Data storage equipment
Cache memory
Data transfer
Printed circuit boards
Printed Circuit Board
Parallel Applications
Data Transfer
Simulators
Latency
Degradation
Simulator
High Performance
Classify
Line
Evaluation

Keywords

  • Command-level simulator
  • Single-chip microprocessor
  • Snoop cache protocol

ASJC Scopus subject areas

  • Hardware and Architecture
  • Information Systems
  • Theoretical Computer Science
  • Computational Theory and Mathematics

Cite this

A study on snoop cache systems for single-chip multiprocessors. / Terasawa, Takuya; Inoue, Keisuke; Kurosawa, Hitoshi; Amano, Hideharu.

In: Systems and Computers in Japan, Vol. 28, No. 2, 02.1998, p. 62-72.

Research output: Contribution to journalArticle

Terasawa, T, Inoue, K, Kurosawa, H & Amano, H 1998, 'A study on snoop cache systems for single-chip multiprocessors', Systems and Computers in Japan, vol. 28, no. 2, pp. 62-72.
Terasawa, Takuya ; Inoue, Keisuke ; Kurosawa, Hitoshi ; Amano, Hideharu. / A study on snoop cache systems for single-chip multiprocessors. In: Systems and Computers in Japan. 1998 ; Vol. 28, No. 2. pp. 62-72.
@article{7e40010da9414117a8e97fb8ca044e3d,
title = "A study on snoop cache systems for single-chip multiprocessors",
abstract = "Progress in device design and implementation technologies will change the structure of bus-connected multiprocessors and allow them to be implemented on a single chip. In such an implementation, the speed of the bus inside the chip is far faster than that of the backplane bus, and data transfer between the chip and external devices will become a bottleneck. Many studies on snoop cache protocols have been made, but they generally assume that the cache memory is on a printed circuit board. In this paper, we first classify the snoop cache protocols in terms of accesses to off-chip shared memory, which will be the principal cause of performance degradation, and then compare them quantitatively. Evaluations are made with an instruction-level multiprocessor simulator and practical parallel applications, varying the cache size or the access latency of shared memory. The results show that an I/N/C protocol that actively uses the line transfers between caches achieves the highest performance under all conditions.",
keywords = "Command-level simulator, Single-chip microprocessor, Snoop cache protocol",
author = "Takuya Terasawa and Keisuke Inoue and Hitoshi Kurosawa and Hideharu Amano",
year = "1998",
month = "2",
language = "English",
volume = "28",
pages = "62--72",
journal = "Systems and Computers in Japan",
issn = "0882-1666",
publisher = "John Wiley and Sons Inc.",
number = "2",

}

TY - JOUR

T1 - A study on snoop cache systems for single-chip multiprocessors

AU - Terasawa, Takuya

AU - Inoue, Keisuke

AU - Kurosawa, Hitoshi

AU - Amano, Hideharu

PY - 1998/2

Y1 - 1998/2

N2 - Progress in device design and implementation technologies will change the structure of bus-connected multiprocessors and allow them to be implemented on a single chip. In such an implementation, the speed of the bus inside the chip is far faster than that of the backplane bus, and data transfer between the chip and external devices will become a bottleneck. Many studies on snoop cache protocols have been made, but they generally assume that the cache memory is on a printed circuit board. In this paper, we first classify the snoop cache protocols in terms of accesses to off-chip shared memory, which will be the principal cause of performance degradation, and then compare them quantitatively. Evaluations are made with an instruction-level multiprocessor simulator and practical parallel applications, varying the cache size or the access latency of shared memory. The results show that an I/N/C protocol that actively uses the line transfers between caches achieves the highest performance under all conditions.

AB - Progress in device design and implementation technologies will change the structure of bus-connected multiprocessors and allow them to be implemented on a single chip. In such an implementation, the speed of the bus inside the chip is far faster than that of the backplane bus, and data transfer between the chip and external devices will become a bottleneck. Many studies on snoop cache protocols have been made, but they generally assume that the cache memory is on a printed circuit board. In this paper, we first classify the snoop cache protocols in terms of accesses to off-chip shared memory, which will be the principal cause of performance degradation, and then compare them quantitatively. Evaluations are made with an instruction-level multiprocessor simulator and practical parallel applications, varying the cache size or the access latency of shared memory. The results show that an I/N/C protocol that actively uses the line transfers between caches achieves the highest performance under all conditions.

KW - Command-level simulator

KW - Single-chip microprocessor

KW - Snoop cache protocol

UR - http://www.scopus.com/inward/record.url?scp=5844313229&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=5844313229&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:5844313229

VL - 28

SP - 62

EP - 72

JO - Systems and Computers in Japan

JF - Systems and Computers in Japan

SN - 0882-1666

IS - 2

ER -