Skip to main content
Log in

The performance of disk arrays in shared-memory database machines

  • Published:
Distributed and Parallel Databases Aims and scope Submit manuscript

Abstract

Disk arrays and shared-memory multiprocessors are new technologies that are rapidly becoming pervasive. They are complementary because disk arrays naturally balance the I/O workload by interleaving data across all disks while a shared-memory multiprocessor balances the processing workload across multiple processors. In this paper, we examine how disk arrays and shared memory multiprocessors lead to an effective method for constructing database machines for general-purpose complex query processing. We show that disk arrays can lead to cost-effective storage systems if they are configured from suitably small formfactor disk drives. We introduce the storage system metricdata temperature (IO/s/Gbyte) as a way to evaluate how well a disk configuration can sustain its workload, and we show that disk arrays can sustain the same data temperature as a more expensive mirrored-disk configuration. We use the metric to evaluate the performance of disk arrays in XPRS, an operational shared-memory multiprocessor database system being developed at the University of California, Berkeley.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  1. A. Bhide and M. Stonebraker, “A performance comparison of two architectures for fast transaction processing,” inProc. 1988 IEEE Data Engineering Conf., Los Angeles, 1988.

  2. D. Bitton, “Arm scheduling in shadowed disks,” inI.E.E.E. Spring COMPCON 1989, San Francisco, 1990, pp. 132–136.

  3. D. Bitton et al., “Benchmarking database systems: a systematic approach,” inProc. 1983 VLDB Conf.

  4. D. Bitton and J. Gray, “Disk shadowing,” inProc. 14th Very Large Database Conf., 1988, pp. 331–338.

  5. P. M. Chen, G.A. Gibson, R.H. Katz, and D.A. Patterson, “An evaluation of redundant arrays of disks using an Amdahl 5890,” inACM SIGMETRICS Conf., Boulder, 1990.

  6. G. Copeland et al., “Data placement in BUBBA,” inProc. 1988 ACM-SIGMOD Conf., Chicago, 1988.

  7. D. Dewitt et al., “GAMMA: a high performance dataflow database machine,” inProc. 1986 VLDB Conf., Kyoto, Japan, 1986.

  8. Fujitsu Product Specification, “M2361A mini-disk drive engineering specifications (revised),” B03P-4825-0001A, 1987.

  9. G.A. Gibson, “Performance and reliability in redundant arrays of inexpensive disks,” inProc. 1989 Annual Computer Measurement Group (CMG) Conf., Reno, NV, 1989.

  10. G.A. Gibson, “Performance and reliability of disk arrays,” Ph.D. dissertation, University of California, Berkeley, 1991.

    Google Scholar 

  11. S. Goldstein, “Storage performance — an eight year outlook,” IBM Santa Theresa Laboratory Technical Report TR 03.308-1, 1987.

  12. J. Gray, R. Horbst, and M. Walker, “Parity striping of disc arrays: low-cost reliable storage with acceptable throughput,” inProc. 16th Very Large Database Conference, Brisbane, Australia, 1990, pp. 148–159.

  13. W. Hong, “Exploiting inter-operation parallelism in XPRS,”Proc. 1992 ACM-SIGMOD Conf. (to appear).

  14. W. Hong and M. Stonebraker, “Optimization of parallel query execution plans in XPRS,” inProc. First Int. Conf. Parallel and Distributed Information Systems, Miami, 1991.

  15. IBM Product Specification, “IBM 3380 direct access storage introduction,” IBM GC 26-4491-0, 1987.

  16. IBM Product Description, 3.5″ Corsair disk drive.

  17. J. Menon and Kasson, “Methods for improved update performance of disk arrays,” inHawaii Int. Conf. System Sciences, Honolulu, 1992.

  18. J. Menon and R. Mattson, “Performance of disk arrays in transaction processing environments,” IBM Research Report, Almaden Research Center, San Jose, CA, 1991.

  19. J. Menon and R. Mattson, “Comparison of sparing alternatives for disk arrays,” inProc. Int. Symp. Computer Architecture, Gold Coast, Australia, 1992.

  20. R. Muntz and K. Lui, “Performance of disk arrays under failure,” inProc. Very Large Database Conf., 1990.

  21. D.A. Patterson, G.A. Gibson, and R.H. Katz, “The case for RAID: redundant arrays of inexpensive disks,” inProc. ACM SIGMOD Conf., Chicago, 1988, pp. 106–113.

  22. M. Rosenblum and J. Ousterhout, “The design and implementation of a log-structured file system,”ACM Trans. Comput. Systems, February, 1992.

  23. K. Salem and H. Garcia-Molina, “Disk striping,” inProc. IEEE Data Engineering Conf., Los Angeles, 1986.

  24. M. Stonebraker, “The case for shared nothing,” inProc. 1986 IEEE Data Engineering Conf., 1986.

  25. M. Stonebraker et al., “The design of XPRS,” inProc. 1988 VLDB Conf., Los Angeles, CA, 1988.

  26. M. Stonebraker et al., “The Postgres next generation DBMS,”Comm. ACM, vol. 34, no. 10, 1991.

  27. J. Tuttle, Personal communication, based on Dataquest Disk Trend Report, 1991.

  28. J. Voelcker, “Winchester disks reach for a gigabyte,”IEEE Spectrum, February, pp. 64–67, 1987.

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Katz, R.H., Hong, W. The performance of disk arrays in shared-memory database machines. Distrib Parallel Databases 1, 167–198 (1993). https://doi.org/10.1007/BF01264050

Download citation

  • Received:

  • Revised:

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF01264050

Keywords

Navigation