Abstract
Several data replication strategies have been proposed to provide high data availability for database applications. However, the trade-offs among the different strategies for various workloads and different operating modes have not been studied before. In this paper, we study the relative performance of three high availability data replication strategies, chained declustering, mirrored disks, and interleaved declustering, in a shared nothing database machine environment. In particular, we have examined (1) the relative performance of the three strategies when no failures have occurred, (2) the effect of load imbalance caused by a disk or processor failure on system throughput and response time, and (3) the tradeoff between the benefit of intra query parallelism and the overhead of activating and scheduling extra operator process. Experimental results obtained from a simulation study indicate that, in the normal mode of operation, chained declustering and interleaved declustering perform comparably. Both perform better than mirrored disks if an application is I/O bound, but slightly worse than mirrored disks if the application is CPU bound. In the event of a disk failure, because chained declustering is able to balance the workload among all remaining operational disks while the other two cannot, it provides noticeably better performance than interleaved declustering and much better performance than mirrored disks.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
D. Bitton, “Arm scheduling in shadowed disks,” inCOMPCON, IEEE Press, 1989.
D. Bitton and J. Gray, “Disk shadowing,” inProc. 14th Int. Conf. Very Large Data Base, Los Angeles, CA, 1988.
A. Borr, “Transaction monitoring in encompass [TM]: Reliable distributed transaction processing,” inProc. 7th Inter. Conf. Very Large Data Base, 1981.
M. Carey and H. Lu, “Load balancing in a locally distributed database system,” inProc. ACM-SIGMOD Int. Conf. Management of Data, 1986.
P. Chen, G. Gibson, K. Katz, and D. Patterson, “An evaluation of redundant arrays of disks using an amdahl 5890,” inProc. ACM SIGMETRICS Conf., 1990.
G. Copeland, W. Alexander, E. Boughter, and T. Keller, “Data placement in bubba,” inProc. ACM-SIGMOD Int. Conf. Management of Data, 1988.
G. Copeland and T. Keller, “A comparison of high-availability media recovery techniques,” inProc. ACM-SIGMOD Int. Conf. Management of Data, 1989.
D. DeWitt, R. Gerber, G. Graefe, M. Heytens, K. Kumar, and M. Muralikrishna, “GAMMA-A high performance dataflow database machine,” inProc. 12th Int. Conf. Very Large Data Base, 1986.
D. DeWitt, S. Ghandeharizadeh, and D. Schneider, “A performance analysis of the gamma database machine,” inProc. ACM-SIGMOD Int. Conf. Management of Data, 1988.
D. DeWitt, S. Ghandeharizadeh, D. Schneider, A. Bricker, H. Hsiao, and R. Rasmussen, “The gamma database machine project,”IEEE Trans. Knowledge Data Eng., vol. 2, no 1, 1990.
P. Frank, “Advances in head technology,” presentation atChallenges in Disk Technology Short Course, Institute for Information Storage Technology, Santa Clara University, Santa Clara, CA, 1987.
J. Gray, B. Horst, and M. Walker, “Parity striping of disk arrays: Low-cost reliable storage with acceptable throughput,” inProc. 16th Int. Conf. Very Large Data Base, 1990.
J. Gray, H. Sammer, and S. Whitford, “Shortest seek vs shortest service time scheduling of mirrored disc reads” Tandem Computers, 1988.
H. Hsiao, “Performance and availability in database machines with replicated data,” Computer Sciences Technical Report#963, University of Wisconsin-Madison, 1990.
H. Hsiao and D. DeWitt, “Chained declustering: A new availabilty strategy for multiprocessor database machines,” inProc. 6th Int. Conf. Data Engineering, Los Angeles, CA, 1990.
Intel Corporation,iPSC/2 User's, Intel Corporation Order No. 311532-002, 1988.
B. Joy, Presentation at ISSCC'85 panel session, Feb. 1985.
J. Katzman, “A fault-tolerant computing system,” inProc. 11th Hawaii Conf. System Sciences, 1978.
M. Kim, “Synchronized disk interleaving,”IEEE Trans. Compu., vol. C-35, no.11, 1986.
M. Livny, “DeNet user's guide,” Version 1.5, Computer Sciences Department, University of Wisconsin-Madison, 1989.
M. Livny, S. Khoshafian, and H. Boral, “Multi-disk management,” inProc. ACM SIGMETRICS Conf., Alberta, Canada, 1987.
D. Patterson, G. Gibson, and R. Katz, “A case for redundant arrays of inexpensive disks(RAID),” inProc. ACM-SIGMOD Int. Conf. Management of Data, Chicago, 1988.
D. Ries and R. Epstein, “Evaluation of distribution criteria for distributed database systems,” UCB/ERL Technical Report M78/22, UC Berkeley, May 1978.
M. Stonebraker, “The case for shared nothing,” inDatabase Eng., vol. 9, no.1, 1986.
M. Stonebraker and G. Schloss, “Distributed RAID-a new multiple copy algorithm,” inProc. 6th Int. Conf. Data Engineering, Los Angeles CA, 1990.
Tandem Database Group, “Nonstop SQL, a distributed high-performance, high-reliability implementation of SQL,” inWorkshop on High Performance Transaction Systems, Asilomar, CA, 1987.
Teradata, “DBC/1012 database computer system manual release 2.0,” Document No. C10-0001-02, Teradata Corp., 1985.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Hsiao, HI., Dewitt, D.J. A performance study of three high availability data replication strategies. Distrib Parallel Databases 1, 53–79 (1993). https://doi.org/10.1007/BF01277520
Issue Date:
DOI: https://doi.org/10.1007/BF01277520