View in EDS

Efficient load balancing in partitioned queries under random perturbations

Saved in:
Bibliographic Details
Title: Efficient load balancing in partitioned queries under random perturbations
Authors: Gounaris, Anastasios, Yfoulis, Christos A., Paton, Norman W.
Source: Gounaris, A, Yfoulis, C A & Paton, N W 2012, 'Efficient load balancing in partitioned queries under random perturbations', ACM Transactions on Autonomous and Adaptive Systems, vol. 7, no. 1, 5. https://doi.org/10.1145/2168260.2168265
Publisher Information: Association for Computing Machinery (ACM), 2012.
Publication Year: 2012
Subject Terms: LQR, Adaptation cost, Control theory, Partitioned queries, 0202 electrical engineering, electronic engineering, information engineering, 02 engineering and technology, Load balancing
Description: This work investigates a particular instance of the problem of designing efficient adaptive systems, under the condition that each adaptation decision incurs some nonnegligible cost when enacted. More specifically, we deal with the problem of dynamic, intraquery load balancing in parallel database queries across heterogeneous nodes in a way that takes into account the inherent cost of adaptations and thus avoids both overreacting and deciding when to adapt in a completely heuristic manner. The latter may lead to serious performance degradation in several cases, such as periodic and random imbalances. We follow a control theoretical approach to this problem; more specifically, we propose a multiple-input multiple-output feedback linear quadratic regulation (LQR) controller, which captures the tradeoff between reaching a balanced state and the cost inherent in such adaptations. Our approach, apart from benefitting from and being characterized by a solid theoretical foundation, exhibits better performance than state-of-the-art heuristics in realistic situations, as verified by thorough evaluation.
Document Type: Article
Language: English
ISSN: 1556-4703
1556-4665
DOI: 10.1145/2168260.2168265
Access URL: http://delab.csd.auth.gr/papers/TAAS11gyp.pdf
https://research.manchester.ac.uk/en/publications/583b47a5-6492-4751-b9ba-dbc779a54824
https://doi.org/10.1145/2168260.2168265
https://dblp.uni-trier.de/db/journals/taas/taas7.html#GounarisYP12
https://www.research.manchester.ac.uk/portal/en/publications/efficient-load-balancing-in-partitioned-queries-under-random-perturbations(583b47a5-6492-4751-b9ba-dbc779a54824).html
https://www.escholar.manchester.ac.uk/uk-ac-man-scw:186852
https://doi.org/10.1145/2168260.2168265
http://www.cs.man.ac.uk/~norm/papers/taas12.pdf
https://dl.acm.org/doi/pdf/10.1145/2168260.2168265
Rights: URL: https://www.acm.org/publications/policies/copyright_policy#Background
Accession Number: edsair.doi.dedup.....f1e55f3ddd0987644e72ab4ce92891eb
Database: OpenAIRE
Description
Abstract:This work investigates a particular instance of the problem of designing efficient adaptive systems, under the condition that each adaptation decision incurs some nonnegligible cost when enacted. More specifically, we deal with the problem of dynamic, intraquery load balancing in parallel database queries across heterogeneous nodes in a way that takes into account the inherent cost of adaptations and thus avoids both overreacting and deciding when to adapt in a completely heuristic manner. The latter may lead to serious performance degradation in several cases, such as periodic and random imbalances. We follow a control theoretical approach to this problem; more specifically, we propose a multiple-input multiple-output feedback linear quadratic regulation (LQR) controller, which captures the tradeoff between reaching a balanced state and the cost inherent in such adaptations. Our approach, apart from benefitting from and being characterized by a solid theoretical foundation, exhibits better performance than state-of-the-art heuristics in realistic situations, as verified by thorough evaluation.
ISSN:15564703
15564665
DOI:10.1145/2168260.2168265