Abstract
The configuration of a distributed database system consists of a network and a number of local configurations. A local configuration (LC) consists of a CPU, some I/O devices and a local DBMS, etc. at its site. To improve the performance of the distributed database system, one way is to change one or more LCs. If, at each site, the old LC may be replaced by one of several possible new LCs, then a large number of different new configurations may be formed, each will have different performance. The objective is to find a new configuration satisfying a certain performance goal at the minimum additional cost. It may not be practical to enumerate all possible configurations and conduct an experiment for each configuration to find the performance since the number of different configurations can be prohibitively large. In this paper, we propose a methodology that predicts the performance of all different configurations based on experiments on a very limited number of configurations. Our experimental results indicate that the proposed methodology fairly accurately predicts the performances of new configurations.
Research supported in part by NSF (IRI-9111988) and AirForce (AFOSR 93-1-0059).
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Agrawal, P., D. Bitton, K. Guh, C. Liu and C. Yu, A Case Study for Distributed Query Processing, Int'l Symposium on Databases in Parallel and Distributed Systems, Austin, Dec. 1988, pp. 124–130.
Dina Bitton, David J. DeWitt and Carolyn Turbyfill, Benchmarking Database Systems: A Systematic Approach. Proc. of VLDB, 1983, pp8–19.
Ceri, S., Negri, M. and Pelagatti, G., Horizontal Partitioning in Database Design, ACM SIGMOD, 1982.
Chen, J.S.J. and Li, Victor, Optimizing Joins in Fragmented Database Systems on a Broadcast Local Network. IEEE TSE, Vol. 15, No.1, Jan. 1989, pp. 26–38.
Epstein, R., Stonebraker, M. and Wong, E., Distributed Query Processing in Relational Database Systems. ACM SIGMOD 1978.
Garcia-Molina, H., Performance Through Memory. Proc. of the ACM SIGMETRICS Conf., 1987, pp. 155–162.
Haas, L.M., Selinger, P.G., and Bertino, E., R +: A Research Project on Distributed Relational DBMS, Database Engineering, Vol. 5, No. 4, Dec. 1982, pp. 66–75.
Jenq, B-C., Kohler, W., and Towsley, D., A Queueing Network Model for a Distributed Database Testbed System. IEEE TSE, Vol.14, No.7, July 1988.
Mackert, L. and Lohman, G., R * Optimizer Validation and Performance Evaluation for Distributed Queries, Proc. of VLDB, Aug. 1986, pp. 149–159.
Liu, C. and Yu, C., Validation and Performance Evaluation of the Partition and Replicate Algorithm, Proc. of Int'l Conf. on Distr. Compu. Sys., 1992, pp. 400–407.
Liu, C. and Yu, C., Performance Issues in Distributed Query Processing. IEEE Trans. on Parallel and Distributed Systems, 1993 (to appear).
Meng, W., Liu, C., Sun, W., and Yu, C., Distributed Database System Performance Prediction. CS-TR-92-18, Dept. of CS, State Univ. of NY at Binghamton, 1992.
Mikkilineni, K.P. and Su, S.Y.W., Evaluation of Relational Join Algorithms in a Pipelined Query Processing Environment. IEEE TSE, Vol.14, No.6, 1988.
Pramanik, Sakti, Performance Analysis of a Database Filter Search Hardware, IEEE Trans. on Computer, Vol. 35, No.12, Dec. 1986, pp. 1077–1082.
Pu, C. and J. M. Wha, Performance Evaluation of Global Reading of Entire Databases, Int'l Symposium on Databases in Parallel and Distributed Systems, Austin, Dec. 1988, pp. 167–176.
Segev, A., Optimizing 2-way Joins in Horizontally Partitioned Database Systems. Computer Journal, Vol. 30, No.5, Oct. 1987, pp. 458–468.
Stonebraker, M. and Neuhold, E., A Distributed Data Base Version of INGRES, Second Berkeley Workshop on Distributed Data Management and Computer Networks, Berkeley, CA, 1977, pp. 19–36.
Yu, C.T., et al., Query Processing in a Fragmented Relational Distributed System: Mermaid. IEEE TSE, Aug. 1985, pp. 795–810.
Yu, C.T., Guh, K. C., Brill, D., Chen., A.L.P., Partition Strategy for Distributed Query Processing in Fast Local Networks, IEEE TSE, June, 1989, pp. 780–793.
Yu, C.T., and Chang, C.C., Distributed Query Processing. ACM Computing Surveys, Vol. 16, No.4, December 1984, pp. 399–433.
Yu, P.S. and Dias, D.M., Performance Analysis of Optimistic Concurrency Control Schemes for Systems with Large Memory, Proc. of SIGMETRICS, 1989, pp. 238.
Yu, C. and Liu, C., Experiences with Distributed Query Processing, Proc. IEEE Data Engineering, Feb. 1990, pp. 192–199.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1993 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Meng, W., Liu, C., Sun, W., Yu, C. (1993). Predict query processing cost in a distributed database system. In: MaÅ™Ãk, V., Lažanský, J., Wagner, R.R. (eds) Database and Expert Systems Applications. DEXA 1993. Lecture Notes in Computer Science, vol 720. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-57234-1_11
Download citation
DOI: https://doi.org/10.1007/3-540-57234-1_11
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-57234-3
Online ISBN: 978-3-540-47982-6
eBook Packages: Springer Book Archive