Utilization of the on-chip L2 cache area in CC-NUMA multiprocessors for applications with a small working set

Sung Woo Jung, Hyong Shik Kim, Chu Shik Jhon

Research output: Contribution to journalArticle

Abstract

In CC-NUMA multiprocessor systems, it is important to reduce the remote memory access time. Based upon the fact that increasing the size of the LRU second-level (L2) cache more than a certain value does not reduce the cache miss rate significantly, in this paper, we propose two split L2 caches to utilize the surplus of the L2 cache. The split L2 caches are composed of a traditional LRU cache and another cache to reduce the remote memory access time. Both work together to reduce total L2 cache miss time by keeping remote (or long-distance) blocks as well as recently used blocks. For another cache, we propose two alternatives: an L2-RVC (Level 2 - Remote Victim Cache) and an L2-DAVC (Level 2 Distance-Aware Victim Cache). The proposed split L2 caches reduce total execution time by up to 27%. It is also found that the proposed split L2 caches outperform the traditional single LRU cache of double size.

Original languageEnglish
Pages (from-to)1617-1624
Number of pages8
JournalIEICE Transactions on Information and Systems
VolumeE87-D
Issue number7
Publication statusPublished - 2004 Jul 1
Externally publishedYes

Fingerprint

Data storage equipment

Keywords

  • Cache replacement policy
  • CC-NUMA multiprocessor
  • Distance-aware cache
  • Interconnection network
  • On-chip cache
  • Remote victim cache

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Information Systems
  • Software

Cite this

Utilization of the on-chip L2 cache area in CC-NUMA multiprocessors for applications with a small working set. / Jung, Sung Woo; Kim, Hyong Shik; Jhon, Chu Shik.

In: IEICE Transactions on Information and Systems, Vol. E87-D, No. 7, 01.07.2004, p. 1617-1624.

Research output: Contribution to journalArticle

@article{4a0b48228c4c4490a81d74640994d094,
title = "Utilization of the on-chip L2 cache area in CC-NUMA multiprocessors for applications with a small working set",
abstract = "In CC-NUMA multiprocessor systems, it is important to reduce the remote memory access time. Based upon the fact that increasing the size of the LRU second-level (L2) cache more than a certain value does not reduce the cache miss rate significantly, in this paper, we propose two split L2 caches to utilize the surplus of the L2 cache. The split L2 caches are composed of a traditional LRU cache and another cache to reduce the remote memory access time. Both work together to reduce total L2 cache miss time by keeping remote (or long-distance) blocks as well as recently used blocks. For another cache, we propose two alternatives: an L2-RVC (Level 2 - Remote Victim Cache) and an L2-DAVC (Level 2 Distance-Aware Victim Cache). The proposed split L2 caches reduce total execution time by up to 27{\%}. It is also found that the proposed split L2 caches outperform the traditional single LRU cache of double size.",
keywords = "Cache replacement policy, CC-NUMA multiprocessor, Distance-aware cache, Interconnection network, On-chip cache, Remote victim cache",
author = "Jung, {Sung Woo} and Kim, {Hyong Shik} and Jhon, {Chu Shik}",
year = "2004",
month = "7",
day = "1",
language = "English",
volume = "E87-D",
pages = "1617--1624",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "7",

}

TY - JOUR

T1 - Utilization of the on-chip L2 cache area in CC-NUMA multiprocessors for applications with a small working set

AU - Jung, Sung Woo

AU - Kim, Hyong Shik

AU - Jhon, Chu Shik

PY - 2004/7/1

Y1 - 2004/7/1

N2 - In CC-NUMA multiprocessor systems, it is important to reduce the remote memory access time. Based upon the fact that increasing the size of the LRU second-level (L2) cache more than a certain value does not reduce the cache miss rate significantly, in this paper, we propose two split L2 caches to utilize the surplus of the L2 cache. The split L2 caches are composed of a traditional LRU cache and another cache to reduce the remote memory access time. Both work together to reduce total L2 cache miss time by keeping remote (or long-distance) blocks as well as recently used blocks. For another cache, we propose two alternatives: an L2-RVC (Level 2 - Remote Victim Cache) and an L2-DAVC (Level 2 Distance-Aware Victim Cache). The proposed split L2 caches reduce total execution time by up to 27%. It is also found that the proposed split L2 caches outperform the traditional single LRU cache of double size.

AB - In CC-NUMA multiprocessor systems, it is important to reduce the remote memory access time. Based upon the fact that increasing the size of the LRU second-level (L2) cache more than a certain value does not reduce the cache miss rate significantly, in this paper, we propose two split L2 caches to utilize the surplus of the L2 cache. The split L2 caches are composed of a traditional LRU cache and another cache to reduce the remote memory access time. Both work together to reduce total L2 cache miss time by keeping remote (or long-distance) blocks as well as recently used blocks. For another cache, we propose two alternatives: an L2-RVC (Level 2 - Remote Victim Cache) and an L2-DAVC (Level 2 Distance-Aware Victim Cache). The proposed split L2 caches reduce total execution time by up to 27%. It is also found that the proposed split L2 caches outperform the traditional single LRU cache of double size.

KW - Cache replacement policy

KW - CC-NUMA multiprocessor

KW - Distance-aware cache

KW - Interconnection network

KW - On-chip cache

KW - Remote victim cache

UR - http://www.scopus.com/inward/record.url?scp=3142765294&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=3142765294&partnerID=8YFLogxK

M3 - Article

VL - E87-D

SP - 1617

EP - 1624

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 7

ER -