The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Les multiprocesseurs à mémoire partagée sont fréquemment utilisés comme serveurs de calcul avec plusieurs programmes parallèles s'exécutant en même temps. Dans de tels environnements, un système d'exploitation change les contextes de plusieurs processus. Lorsque le système d'exploitation change de contexte, outre le coût de sauvegarde du contexte du processus en cours de remplacement et celui de l'intégration du contexte du nouveau processus à exécuter, les performances du cache des processeurs peuvent également être affectées. L'algorithme bloqué améliore les performances du cache en augmentant la localité des références mémoire. Dans un programme bloqué utilisant cet algorithme, les performances du programme peuvent être considérablement affectées par la réutilisation d'un bloc chargé dans une mémoire cache. Si des changements de contexte fréquents remplacent le bloc avant qu'il ne soit complètement réutilisé, la localité du cache dans un programme bloqué ne peut pas être exploitée avec succès. Pour résoudre ce problème, nous proposons une politique sécurisée par préemption pour utiliser la localité du cache des programmes bloqués dans un système multiprogrammé. La politique proposée retarde le changement de contexte jusqu'à ce qu'un bloc soit entièrement réutilisé dans un programme, mais compense également le temps processeur monopolisé sur les mécanismes de planification du processeur. Nos résultats de simulation montrent que dans une situation où des programmes bloqués sont exécutés sur des multiprocesseurs multiprogrammés à mémoire partagée, la politique proposée améliore les performances de ces programmes en raison d'une diminution des échecs de cache. Dans de telles situations, cela a également un impact bénéfique sur les performances globales du système en raison de l’utilisation améliorée du processeur.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Inbum JUNG, Jongwoong HYUN, Joonwon LEE, "A Scheduling Policy for Blocked Programs in Multiprogrammed Shared-Memory Multiprocessors" in IEICE TRANSACTIONS on Information,
vol. E83-D, no. 9, pp. 1762-1771, September 2000, doi: .
Abstract: Shared memory multiprocessors are frequently used as compute servers with multiple parallel programs executing at the same time. In such environments, an operating system switches the contexts of multiple processes. When the operating system switches contexts, in addition to the cost of saving the context of the process being swapped out and that of bringing in the context of the new process to be run, the cache performance of processors also can be affected. The blocked algorithm improves cache performance by increasing the locality of memory references. In a blocked program using this algorithm, program performance can be significantly affected by the reuse of a block loaded into a cache memory. If frequent context switching replaces the block before it is completely reused, the cache locality in a blocked program cannot be successfully exploited. To address this problem, we propose a preemption-safe policy to utilize the cache locality of blocked programs in a multiprogrammed system. The proposed policy delays context switching until a block is fully reused within a program, but also compensates for the monopolized processor time on processor scheduling mechanisms. Our simulation results show that in a situation where blocked programs are run on multiprogrammed shared-memory multiprocessors, the proposed policy improves the performance of these programs due to a decrease in cache misses. In such situations, it also has a beneficial impact on the overall system performance due to the enhanced processor utilization.
URL: https://global.ieice.org/en_transactions/information/10.1587/e83-d_9_1762/_p
Copier
@ARTICLE{e83-d_9_1762,
author={Inbum JUNG, Jongwoong HYUN, Joonwon LEE, },
journal={IEICE TRANSACTIONS on Information},
title={A Scheduling Policy for Blocked Programs in Multiprogrammed Shared-Memory Multiprocessors},
year={2000},
volume={E83-D},
number={9},
pages={1762-1771},
abstract={Shared memory multiprocessors are frequently used as compute servers with multiple parallel programs executing at the same time. In such environments, an operating system switches the contexts of multiple processes. When the operating system switches contexts, in addition to the cost of saving the context of the process being swapped out and that of bringing in the context of the new process to be run, the cache performance of processors also can be affected. The blocked algorithm improves cache performance by increasing the locality of memory references. In a blocked program using this algorithm, program performance can be significantly affected by the reuse of a block loaded into a cache memory. If frequent context switching replaces the block before it is completely reused, the cache locality in a blocked program cannot be successfully exploited. To address this problem, we propose a preemption-safe policy to utilize the cache locality of blocked programs in a multiprogrammed system. The proposed policy delays context switching until a block is fully reused within a program, but also compensates for the monopolized processor time on processor scheduling mechanisms. Our simulation results show that in a situation where blocked programs are run on multiprogrammed shared-memory multiprocessors, the proposed policy improves the performance of these programs due to a decrease in cache misses. In such situations, it also has a beneficial impact on the overall system performance due to the enhanced processor utilization.},
keywords={},
doi={},
ISSN={},
month={September},}
Copier
TY - JOUR
TI - A Scheduling Policy for Blocked Programs in Multiprogrammed Shared-Memory Multiprocessors
T2 - IEICE TRANSACTIONS on Information
SP - 1762
EP - 1771
AU - Inbum JUNG
AU - Jongwoong HYUN
AU - Joonwon LEE
PY - 2000
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E83-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2000
AB - Shared memory multiprocessors are frequently used as compute servers with multiple parallel programs executing at the same time. In such environments, an operating system switches the contexts of multiple processes. When the operating system switches contexts, in addition to the cost of saving the context of the process being swapped out and that of bringing in the context of the new process to be run, the cache performance of processors also can be affected. The blocked algorithm improves cache performance by increasing the locality of memory references. In a blocked program using this algorithm, program performance can be significantly affected by the reuse of a block loaded into a cache memory. If frequent context switching replaces the block before it is completely reused, the cache locality in a blocked program cannot be successfully exploited. To address this problem, we propose a preemption-safe policy to utilize the cache locality of blocked programs in a multiprogrammed system. The proposed policy delays context switching until a block is fully reused within a program, but also compensates for the monopolized processor time on processor scheduling mechanisms. Our simulation results show that in a situation where blocked programs are run on multiprogrammed shared-memory multiprocessors, the proposed policy improves the performance of these programs due to a decrease in cache misses. In such situations, it also has a beneficial impact on the overall system performance due to the enhanced processor utilization.
ER -