Home-cluster based processor scheduling with page placement for NUMA multiprocessors

被引：0

作者：

Koita, T ^{[1
]}

Katayama, T ^{[1
]}

Saisho, K ^{[1
]}

Fukuda, A ^{[1
]}

机构：

[1] Nara Inst Sci & Technol, Nara 6300101, Japan

来源：

INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES AND APPLICATIONS, VOLS I-V, PROCEEDINGS | 1999年

关键词：

processor scheduling; dynamic space-sharing; page placement; NUMA multiprocessors; multiprogrammed environments;

D O I：

暂无

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper proposes a home-cluster based policy for cluster-based NUMA (Non-Uniform Memory Access) multiprocessors. The home-cluster is a reserved cluster for processor and memory page allocation to a process. The policy consists of a processor scheduling scheme and a page placement scheme. Both schemes work together based on the concept of the home-cluster. The policy focuses on not only the processor scheduling scheme but also the page placement scheme. Several alternatives of the home-cluster based policy are examined. And then, the home-cluster based policy is compared with other policies which are the combinations of an existing processor scheduling scheme and a conventional page placement scheme. The simulation results show that the proposed policy performs better than the others because it can reduce remote access overhead and mean response time of a process.

引用

页码：539 / 545

页数：7

共 13 条

[1]

BOLOSKY WJ, 1991, SIGPLAN NOTICES, V26, P212, DOI 10.1145/106973.106994

[2]

Feitelson D. G., 1994, SURVEY SCHEDULING MU

[3]

FUKUDA A, 1993, P INT COMP SOFTW APP, P343, DOI 10.1109/CMPSAC.1993.404239

[4]

GUPTA A, 1991, P 1991 ACM SIGMETRIC, P120

[5]

KAI H, 1994, P 6 JOINT S PAR PROC, P319

[6]

Koita T., 1998, Proceedings of the 16th IASTED International Conference. Applied Informatics, P265

[7]

Koita T, 1998, INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES AND APPLICATIONS, VOLS I-IV, PROCEEDINGS, P851

[8] EXPERIMENTAL COMPARISON OF MEMORY MANAGEMENT POLICIES FOR NUMA MULTIPROCESSORS [J].

LAROWE, RP ;

ELLIS, CS .

ACM TRANSACTIONS ON COMPUTER SYSTEMS, 1991, 9 (04) :319-363

[9]

LAROWE RP, 1991, P 13 ACM S OP SYST P, P137

[10] A DYNAMIC PROCESSOR ALLOCATION POLICY FOR MULTIPROGRAMMED SHARED-MEMORY MULTIPROCESSORS [J].

MCCANN, C ;

VASWANI, R ;

ZAHORJAN, J .

ACM TRANSACTIONS ON COMPUTER SYSTEMS, 1993, 11 (02) :146-178

← 1 2 →