Revisiting Shared Cache Contention Problems: A Practical Hardware-Software Cooperative Approach

Eunji PAK; Sang-Hoon KIM; Jaehyuk HUH; Seungryoul MAENG

doi:10.1587/transinf.E96.D.1457

Revisiting Shared Cache Contention Problems: A Practical Hardware-Software Cooperative Approach

Eunji PAK, Sang-Hoon KIM, Jaehyuk HUH, Seungryoul MAENG

Full Text Views

0

Share
Cite this

Summary :

Although shared caches allow the dynamic allocation of limited cache capacity among cores, traditional LRU replacement policies often cannot prevent negative interference among cores. To address the contention problem in shared caches, cache partitioning and application scheduling techniques have been extensively studied. Partitioning explicitly determines cache capacity for each core to maximize the overall throughput. On the other hand, application scheduling by operating systems groups the least interfering applications for each shared cache, when multiple shared caches exist in systems. Although application scheduling can mitigate the contention problem without any extra hardware support, its effect can be limited for some severe contentions. This paper proposes a low cost solution, based on application scheduling with a simple cache insertion control. Instead of using a full hardware-based cache partitioning mechanism, the proposed technique mostly relies on application scheduling. It selectively uses LRU insertion to the shared caches, which can be added with negligible hardware changes from the current commercial processor designs. For the completeness of cache interference evaluation, this paper examines all possible mixes from a set of applications, instead of using a just few selected mixes. The evaluation shows that the proposed technique can mitigate the cache contention problem effectively, close to the ideal scheduling and partitioning.

Publication: IEICE TRANSACTIONS on Information Vol.E96-D No.7 pp.1457-1466

Publication Date: 2013/07/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E96.D.1457

Type of Manuscript: PAPER

Category: Computer System

Cite this

Copy

Eunji PAK, Sang-Hoon KIM, Jaehyuk HUH, Seungryoul MAENG, "Revisiting Shared Cache Contention Problems: A Practical Hardware-Software Cooperative Approach" in IEICE TRANSACTIONS on Information, vol. E96-D, no. 7, pp. 1457-1466, July 2013, doi: 10.1587/transinf.E96.D.1457.
Abstract: Although shared caches allow the dynamic allocation of limited cache capacity among cores, traditional LRU replacement policies often cannot prevent negative interference among cores. To address the contention problem in shared caches, cache partitioning and application scheduling techniques have been extensively studied. Partitioning explicitly determines cache capacity for each core to maximize the overall throughput. On the other hand, application scheduling by operating systems groups the least interfering applications for each shared cache, when multiple shared caches exist in systems. Although application scheduling can mitigate the contention problem without any extra hardware support, its effect can be limited for some severe contentions. This paper proposes a low cost solution, based on application scheduling with a simple cache insertion control. Instead of using a full hardware-based cache partitioning mechanism, the proposed technique mostly relies on application scheduling. It selectively uses LRU insertion to the shared caches, which can be added with negligible hardware changes from the current commercial processor designs. For the completeness of cache interference evaluation, this paper examines all possible mixes from a set of applications, instead of using a just few selected mixes. The evaluation shows that the proposed technique can mitigate the cache contention problem effectively, close to the ideal scheduling and partitioning.
URL: https://globals.ieice.org/en_transactions/information/10.1587/transinf.E96.D.1457/_p

Copy

@ARTICLE{e96-d_7_1457,
author={Eunji PAK, Sang-Hoon KIM, Jaehyuk HUH, Seungryoul MAENG, },
journal={IEICE TRANSACTIONS on Information},
title={Revisiting Shared Cache Contention Problems: A Practical Hardware-Software Cooperative Approach},
year={2013},
volume={E96-D},
number={7},
pages={1457-1466},
abstract={Although shared caches allow the dynamic allocation of limited cache capacity among cores, traditional LRU replacement policies often cannot prevent negative interference among cores. To address the contention problem in shared caches, cache partitioning and application scheduling techniques have been extensively studied. Partitioning explicitly determines cache capacity for each core to maximize the overall throughput. On the other hand, application scheduling by operating systems groups the least interfering applications for each shared cache, when multiple shared caches exist in systems. Although application scheduling can mitigate the contention problem without any extra hardware support, its effect can be limited for some severe contentions. This paper proposes a low cost solution, based on application scheduling with a simple cache insertion control. Instead of using a full hardware-based cache partitioning mechanism, the proposed technique mostly relies on application scheduling. It selectively uses LRU insertion to the shared caches, which can be added with negligible hardware changes from the current commercial processor designs. For the completeness of cache interference evaluation, this paper examines all possible mixes from a set of applications, instead of using a just few selected mixes. The evaluation shows that the proposed technique can mitigate the cache contention problem effectively, close to the ideal scheduling and partitioning.},
keywords={},
doi={10.1587/transinf.E96.D.1457},
ISSN={1745-1361},
month={July},}

Copy

TY - JOUR
TI - Revisiting Shared Cache Contention Problems: A Practical Hardware-Software Cooperative Approach
T2 - IEICE TRANSACTIONS on Information
SP - 1457
EP - 1466
AU - Eunji PAK
AU - Sang-Hoon KIM
AU - Jaehyuk HUH
AU - Seungryoul MAENG
PY - 2013
DO - 10.1587/transinf.E96.D.1457
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E96-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2013
AB - Although shared caches allow the dynamic allocation of limited cache capacity among cores, traditional LRU replacement policies often cannot prevent negative interference among cores. To address the contention problem in shared caches, cache partitioning and application scheduling techniques have been extensively studied. Partitioning explicitly determines cache capacity for each core to maximize the overall throughput. On the other hand, application scheduling by operating systems groups the least interfering applications for each shared cache, when multiple shared caches exist in systems. Although application scheduling can mitigate the contention problem without any extra hardware support, its effect can be limited for some severe contentions. This paper proposes a low cost solution, based on application scheduling with a simple cache insertion control. Instead of using a full hardware-based cache partitioning mechanism, the proposed technique mostly relies on application scheduling. It selectively uses LRU insertion to the shared caches, which can be added with negligible hardware changes from the current commercial processor designs. For the completeness of cache interference evaluation, this paper examines all possible mixes from a set of applications, instead of using a just few selected mixes. The evaluation shows that the proposed technique can mitigate the cache contention problem effectively, close to the ideal scheduling and partitioning.
ER -