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
La sérialisabilité Epsilon (ESR) a été proposée pour assouplir les contraintes de sérialisabilité en permettant aux transactions de s'exécuter avec un nombre limité d'incohérences (ε-spec). Les algorithmes de contrôle de divergence, considérés comme des extensions des algorithmes de contrôle de concurrence, permettent aux transactions en lecture seule de se terminer si leurs incohérences ne dépassent pas ε-spec. Cet article étudie les performances des algorithmes de contrôle de divergence à verrouillage biphasé (2PLDC) et de contrôle de divergence optimiste (ODC). Nous développons une partie centrale du système de traitement des transactions ESR qui fonctionne avec 2PLDC et ODC. Nous avons appliqué un modèle de simulation de base de données centralisé complet pour mesurer les performances. Les évaluations sont menées avec des charges de travail multi-classes où les transactions de mise à jour en ligne et les requêtes de longue durée progressent sous diverses ε-spec. Nos résultats démontrent que des améliorations significatives des performances sont obtenues avec une incohérence tolérable non nulle. Avec suffisamment de spécifications ε et des ressources système limitées, les deux algorithmes aboutissent à des performances comparables. Cependant, avec un faible conflit de ressources, ODC fonctionne nettement mieux que 2PLDC. De plus, dans la gamme des petites ε-spec, les requêtes validées par ODC ont des résultats plus précis que celles validées par 2PLDC.
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
Akira KAWAGUCHI, Kui W. MOK, Calton PU, Kun-Lung WU, Philip S. YU, "A Performance Study of Divergence Control Algorithms" in IEICE TRANSACTIONS on Information,
vol. E82-D, no. 1, pp. 224-235, January 1999, doi: .
Abstract: Epsilon serializability (ESR) was proposed to relax serializability constraints by allowing transactions to execute with a limited amount of inconsistency (ε-spec). Divergence control algorithms, viewed as extensions of concurrency control algorithms, enable read-only transactions to complete if their inconsistencies do not exceed ε-spec. This paper studies the performance of two-phase locking divergence control (2PLDC) and optimistic divergence control (ODC) algorithms. We develop a central part of the ESR transaction processing system that runs with 2PLDC and ODC. We applied a comprehensive centralized database simulation model to measure the performance. Evaluations are conducted with multi-class workloads where on-line update transactions and long-duration queries progress under various ε-spec. Our results demonstrate that significant performance enhancements are achieved with a non-zero tolerable inconsistency. With sufficient ε-spec and limited system resources, both algorithms result in comparable performance. However, with low resource contention, ODC performs significantly better than 2PLDC. Furthermore, in the range of small ε-spec, the queries committed by ODC have more accurate results than those committed by 2PLDC.
URL: https://global.ieice.org/en_transactions/information/10.1587/e82-d_1_224/_p
Copier
@ARTICLE{e82-d_1_224,
author={Akira KAWAGUCHI, Kui W. MOK, Calton PU, Kun-Lung WU, Philip S. YU, },
journal={IEICE TRANSACTIONS on Information},
title={A Performance Study of Divergence Control Algorithms},
year={1999},
volume={E82-D},
number={1},
pages={224-235},
abstract={Epsilon serializability (ESR) was proposed to relax serializability constraints by allowing transactions to execute with a limited amount of inconsistency (ε-spec). Divergence control algorithms, viewed as extensions of concurrency control algorithms, enable read-only transactions to complete if their inconsistencies do not exceed ε-spec. This paper studies the performance of two-phase locking divergence control (2PLDC) and optimistic divergence control (ODC) algorithms. We develop a central part of the ESR transaction processing system that runs with 2PLDC and ODC. We applied a comprehensive centralized database simulation model to measure the performance. Evaluations are conducted with multi-class workloads where on-line update transactions and long-duration queries progress under various ε-spec. Our results demonstrate that significant performance enhancements are achieved with a non-zero tolerable inconsistency. With sufficient ε-spec and limited system resources, both algorithms result in comparable performance. However, with low resource contention, ODC performs significantly better than 2PLDC. Furthermore, in the range of small ε-spec, the queries committed by ODC have more accurate results than those committed by 2PLDC.},
keywords={},
doi={},
ISSN={},
month={January},}
Copier
TY - JOUR
TI - A Performance Study of Divergence Control Algorithms
T2 - IEICE TRANSACTIONS on Information
SP - 224
EP - 235
AU - Akira KAWAGUCHI
AU - Kui W. MOK
AU - Calton PU
AU - Kun-Lung WU
AU - Philip S. YU
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E82-D
IS - 1
JA - IEICE TRANSACTIONS on Information
Y1 - January 1999
AB - Epsilon serializability (ESR) was proposed to relax serializability constraints by allowing transactions to execute with a limited amount of inconsistency (ε-spec). Divergence control algorithms, viewed as extensions of concurrency control algorithms, enable read-only transactions to complete if their inconsistencies do not exceed ε-spec. This paper studies the performance of two-phase locking divergence control (2PLDC) and optimistic divergence control (ODC) algorithms. We develop a central part of the ESR transaction processing system that runs with 2PLDC and ODC. We applied a comprehensive centralized database simulation model to measure the performance. Evaluations are conducted with multi-class workloads where on-line update transactions and long-duration queries progress under various ε-spec. Our results demonstrate that significant performance enhancements are achieved with a non-zero tolerable inconsistency. With sufficient ε-spec and limited system resources, both algorithms result in comparable performance. However, with low resource contention, ODC performs significantly better than 2PLDC. Furthermore, in the range of small ε-spec, the queries committed by ODC have more accurate results than those committed by 2PLDC.
ER -