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.
- Aström, K. J. and Wittenmark, B. 1995. Adaptive Control. Addison-Wesley, Reading, MA.Google Scholar
- Balazinska, M., Balakrishnan, H., and Stonebraker, M. 2004. Contract-based load management in federated distributed systems. In Proceedings of the ACM Symposium on Networked Systems Design and Implementation (NSDI). 197--210. Google Scholar
Digital Library
- Birdwell, J., Zhong, T., Chiasson, J., Abdallah, C., and Hayat, M. 2006. Resource-constrained load balancing controller for a parallel database. In Proceedings of the American Control Conference.Google Scholar
- Camponogara, E., Jia, D., Krogh, B., and Talukdar, S. 2002. Distributed model predictive control. IEEE Control Syst. Mag. 22, 1, 44--52.Google Scholar
Cross Ref
- Deshpande, A., Ives, Z. G., and Raman, V. 2007. Adaptive query processing. Found. Trends Datab. 1, 1, 1--140. Google Scholar
Digital Library
- DeWitt, D. 1992. Parallel database systems: The future of high performance database systems. Comm. ACM 35, 6, 85--98. Google Scholar
Digital Library
- Diao, Y., Hellerstein, J. L., Parekh, S. S., Griffith, R., Kaiser, G. E., and Phung, D. B. 2005a. Self-managing systems: A control theory foundation. Proc. IEEE, 441--448. Google Scholar
Digital Library
- Diao, Y., Hellerstein, J. L., Storm, A. J., Surendra, M., Lightstone, S., Parekh, S. S., and Garcia-Arellano, C. 2004. Incorporating cost of control into the design of a load balancing controller. In Proceedings ofthe IEEE Real-Time and Embedded Technology and Applications Symposium. 376--387. Google Scholar
Digital Library
- Diao, Y., Wu, C. W., Hellerstein, J. L., Storm, A. J., Surendra, M., Lightstone, S., Parekh, S., Garcia-Arellano, C., Carroll, M., Chu, L., and Colaco, J. 2005b. Comparative studies of load balancing with control and optimization techniques. In Proceedings of the American Control Conference. 1484--1490.Google Scholar
- Dumont, G. and Huzmezan, M. 2002. Concepts, methods and techniques in adaptive control. In Proceedings of the American Control Conference.Google Scholar
- Franklin, G., Powell, J., and Workman, M. 1998. Digital Control of Dynamic Systems 3rd Ed. Addison-Wesley, Reading, MA. Google Scholar
Digital Library
- Fu, Y., Wang, H., Lu, C., and Chandra, R. S. 2006. Distributed utilization control for real-time clusters with load balancing. In Proceedings of the 27th IEEE International Real-Time Systems Symposium (RTSS). 137--146. Google Scholar
Digital Library
- Gedik, B. and Liu, L. 2003. Peercq: A decentralized and self-configuring peer-to-peer information monitoring system. In Proceedings of the IEEE International Conference on Distributed Computer Systems (ICDCS). 490--499. Google Scholar
Digital Library
- Gounaris, A., Yfoulis, C., Sakellariou, R., and Dikaiakos, M. D. 2008a. A control theoretical approach to self-optimizing block transfer in web service grids. ACM Trans. Auton. Adapt. Syst. 3, 2. Google Scholar
Digital Library
- Gounaris, A., Yfoulis, C., Sakellariou, R., and Dikaiakos, M. D. 2008b. Robust runtime optimization of data transfer in queries over web services. In Proceedings of the International Conference on Data Engineering (ICDE). Google Scholar
Digital Library
- Gounaris, A., Yfoulis, C. A., and Paton, N. W. 2009. An efficient load balancing LQR controller in parallel databases queries under random perturbations. In Proceedings ofthe 3rd IEEE Multi-Conference on Systems and Control (2009).Google Scholar
- Graefe, G. 1990. Encapsulation of parallelism in the volcano query processing system. In Proceedings of the ACM SIGMOD International Conference on Management of Data. H. Garcia-Molina and H. V. Jagadish, Eds., 102--111. Google Scholar
Digital Library
- Graefe, G. 1996. Iterators, schedulers, and distributed memory parallelism. Softw. Pract. Exper. 26, 4, 427--452. Google Scholar
Digital Library
- Hellerstein, J. L., Diao, Y., Parekh, S., and Tilbury, D. M. 2004. Feedback Control of Computing Systems. John Wiley & Sons. Google Scholar
Digital Library
- Ives, Z., Florescu, D., Friedman, M., Levy, A., and Weld, D. 1999. An adaptive query execution system for data integration. In Proceedings of the ACM SIGMOD International Conference on Management of Data. 299--310. Google Scholar
Digital Library
- Kephart, J. and Chess, D. 2003. The vision of autonomic computing. IEEE Computer 36, 1, 41--50. Google Scholar
Digital Library
- Kephart, J. and Das, R. 2007. Achieving self-management via utility functions. IEEE Internet Comput. 11, 1, 40--48. Google Scholar
Digital Library
- Lightstone, S., Schiefer, B., Zilio, D., and Kleewein, J. 2003. Autonomic computing for relational databases: The ten-year vision. In Proceedings of the IEEE Workshop Autonomic Computing Principles and Architectures (AUCOPA). 419--424.Google Scholar
- Lightstone, S., Surendra, M., Diao, Y., Parekh, S. S., Hellerstein, J. L., Rose, K., Storm, A. J., and Garcia-Arellano, C. 2007. Control theory: A foundational technique for self managing databases. In Proceedings of the International Conference on Data Engineering Workshops (ICDE). 395--403. Google Scholar
Digital Library
- Lu, H. and Tan, K.-L. 1992. Dynamic and load-balanced task-oriented datbase query processing in parallel systems. In Proceedings ofthe 3rd International Conference on Extending Database Technology (EDBT). 357--372. Google Scholar
Digital Library
- Markl, V., Raman, V., Simmen, D., Lohman, G., and Pirahesh, H. 2004. Robust query processing through progressive optimization. In Proceedings of the ACM SIGMOD International Conference on Management of Data. 659--670. Google Scholar
Digital Library
- Mehta, M. and DeWitt, D. J. 1995. Managing intra-operator parallelism in parallel database systems. In Proceedings of the 21th International Conference on Very Large Data Bases (VLDB). 382--394. Google Scholar
Digital Library
- Ng, K. W., Wang, Z., Muntz, R. R., and Nittel, S. 1999. Dynamic query re-optimization. In Proceedings of the International Conference on Statistical and Scientific Database Management (SSDBM). 264--273. Google Scholar
Digital Library
- Paton, N. W., Chávez, J. B., Chen, M., Raman, V., Swart, G., Narang, I., Yellin, D. M., and Fernandes, A. A. A. 2009. Autonomic query parallelization using non-dedicated computers: an evaluation of adaptivity options. VLDB J. 18, 1, 119--140. Google Scholar
Digital Library
- Rahm, E. and Marek, R. 1995. Dynamic multi-resource load balancing in parallel database systems. In Proceedings of the International Conference on Very Large Data Bases (VLDB). 395--406. Google Scholar
Digital Library
- Raman, V., Han, W., and Narang, I. 2005. Parallel querying with non-dedicated computers. In Proceedings of the International Conference on Very Large Data Bases (VLDB). 61--72. Google Scholar
Digital Library
- Sampaio, S., Paton, N., Smith, J., and Watson, P. 2006. Measuring and Modelling the Performance of a Parallel ODMG Compliant Object Database Server. Concurr. Computat. Pract. Exper. 18, 1, 63--109. Google Scholar
Digital Library
- Shah, M. A., Hellerstein, J. M., Chandrasekaran, S., and Franklin, M. J. 2003. Flux: An adaptive partitioning operator for continuous query systems. In Proceedings of the International Conference on Data Engineering Workshops (ICDE). 25--36.Google Scholar
- Wolf, J. L., Yu, P. S., Turek, J., and Dias, D. M. 1993. A parallel hash join algorithm for managing data skew. IEEE Trans. Parall. Distrib. Syst. 4, 12, 1355--1371. Google Scholar
Digital Library
- Xing, Y., Zdonik, S. B., and Hwang, J.-H. 2005. Dynamic load distribution in the borealis stream processor. In Proceedings of the International Conference on Data Engineering Workshops (ICDE). 791--802. Google Scholar
Digital Library
Index Terms
Efficient load balancing in partitioned queries under random perturbations
Recommendations
Research on Control Method of Two-wheeled Self-balancing Robot
ICICTA '11: Proceedings of the 2011 Fourth International Conference on Intelligent Computation Technology and Automation - Volume 01Based on Newton dynamics mechanics theory make a study of two-wheeled self-balancing robot, a detailed mathematical model of the modeling process is provided, and then, using the reasonable method, a linear state-space equations is built up. After that, ...
Modeling and Simulation for Self-Balance System
ICDMA '10: Proceedings of the 2010 International Conference on Digital Manufacturing & Automation - Volume 01Self-balance system is a kind of special wheeled mobile robots, a typical device used to test control methods. In this paper, based on mechanical analysis, linear and nonlinear models of self-balance system were established. Controllers were designed, ...
Load balancing on a hypercube
IPPS '91: Proceedings of the Fifth International Parallel Processing SymposiumAssuming there is a hypercube with p=2/sup d/ processors and that each hypercube processor, PE(i), has L/sub i/ units of load. The load is always in integral units and may be distributed to other processors in integral units. In the load balancing ...






Comments