Skip to main content

Heterogeneous I/O Contention in a Single-Bus Multiprocessor

  • Chapter
Input/Output in Parallel and Distributed Computer Systems

Abstract

In this article we describe an approach to measuring the performance degradation produced by contention for a single I/O bus by accesses to multiple heterogeneous I/O device types on a multiprocessor system. By heterogeneous we mean device types that have widely different characteristics, such as video, network, and disk devices, as opposed to systems with multiple devices of the same type, such as multiple disks, which we will call homogeneous. We use a correlation analysis to characterize the contention and then quantify the loss in performance by approximating the overhead time of contention using actual trace data.

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

Access this chapter

Chapter
USD 29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD 169.00
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD 219.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info
Hardcover Book
USD 219.99
Price excludes VAT (USA)
  • Durable hardcover edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. G. Biagini and M. Goldfeder. Modeling I/O subsystems concurrency by applying a generalized machine repair model. In G. Iazeolla et al., editors, Computer Performance and Reliability: Proceedings of the 2nd International MCPR Workshop, pages 401–15, 1988.

    Google Scholar 

  2. B. L. Bodnar and A. C. Liu. Modeling and performance analysis of single-bus tightly-coupled multiprocessors. IEEE Transactions on Computers, 38(3): 464–70, 1989.

    Article  Google Scholar 

  3. P. Chen and D. Patterson. A new approach to I/O performance evaluation — self-scaling I/O benchmarks, predicted I/O performance. ACM Transactions on Computer Systems, 12(4):308–39, November 1994.

    Article  Google Scholar 

  4. D. Clark, P. Bannon, and J. Keller. Measuring VAX 8800 performance with a histogram hardware monitor. In 15th Annual International Symposium on Computer Architecture, pages 176–85, 1988.

    Google Scholar 

  5. G. Ganger and Y. Patt. The process-flow model: Examining I/O performance from the system’s point of view. In ACM SIGMETRICS 1993, pages 86–97, 1993.

    Chapter  Google Scholar 

  6. D. Greenberg and A. Shum. Performance analysis of I/O subchannel skews. In CMG 86 International Conference on Management and Performance Evaluation of Computer Systems, Conference Proceedings Las Vegas, NV, pages 189–94, 9–12 Dec. 1986.

    Google Scholar 

  7. J. Harden, D. Reese, F. To, C. Bordiert, and G. Jones. A performance monitor for the MSPARC multicomputer. In Proceedings of IEEE SOUTHEASTCON’92, pages 724–9, 1992.

    Chapter  Google Scholar 

  8. P. Heidelberger and S. Lavenberg. Computer performance evaluation methodology. IEEE Transactions on Computers, C-33(12): 1195–220, December 1984.

    Article  MathSciNet  Google Scholar 

  9. P. Lim and J. Tien. Direct access storage device (DASD) modeling and validation. In IEEE International Conference on Systems, Man, and Cybernetics, Conference Proceedings, pages 1024–9, 14–17 Nov. 1989, Cambridge, MA 1989.

    Chapter  Google Scholar 

  10. S. Ng. Improving disk performance via latency reduction. IEEE Transactions on Computers, 40(1):22–30, January 1991.

    Article  Google Scholar 

  11. I. Onyuksel and K. Irani. Markovian queuing network models for performance analysis of a single-bus multiprocessor system. IEEE Transactions on Computers, 39(7):975–80, July 1990.

    Article  MathSciNet  Google Scholar 

  12. J. Ousterhout et al. A trace-driven analysis of the UNIX 4.2 BSD file system. In Proceedings of the 10th Symposium on Operating System Principles, pages 15–24, December 1985.

    Chapter  Google Scholar 

  13. S. H. VanderLeest and R. K. Iyer. Priority: an I/O instrumentation, measurement, and analysis tool. In Proceedings of Eighth ISCA Conference on Parallel and Distributed Computing Systems (PDCS-95), pages 534–539, 1995.

    Google Scholar 

  14. N. Wilhelm. A general model for the performance of disk systems. Journal of the ACM, 24(1): 14–31, January 1977.

    Article  MathSciNet  MATH  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Editors and Affiliations

Rights and permissions

Reprints and permissions

Copyright information

© 1996 Kluwer Academic Publishers

About this chapter

Cite this chapter

VanderLeest, S.H., Iyer, R.K. (1996). Heterogeneous I/O Contention in a Single-Bus Multiprocessor. In: Jain, R., Werth, J., Browne, J.C. (eds) Input/Output in Parallel and Distributed Computer Systems. The Kluwer International Series in Engineering and Computer Science, vol 362. Springer, Boston, MA. https://doi.org/10.1007/978-1-4613-1401-1_14

Download citation

  • DOI: https://doi.org/10.1007/978-1-4613-1401-1_14

  • Publisher Name: Springer, Boston, MA

  • Print ISBN: 978-1-4612-8607-3

  • Online ISBN: 978-1-4613-1401-1

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics