Skip to main content

Grouped RAID accesses to reduce false sharing effect in clusters with single I/O space

  • VII Poster Session Papers
  • Conference paper
  • First Online:
High Performance Computing (ISHPC 1999)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 1615))

Included in the following conference series:

  • 109 Accesses

Abstract

Disk array technology has been widely used for parallel I/O operations in the array. Distributed RAID can be configured as single I/O space to facilitate programming and performance efficiency in workstation clusters. In single I/O space, the basic operation unit to the disks in the array is stripes. False sharing problem in single I/O space, has been recognized as the major factor to the degradation of overall system performance. In this paper, we present an efficient Length-variable Command Recombination (LCR) method. It reduces the I/O access time without increasing the irrelevant content in the buffer, reducing the false sharing effect. Theoretical analysis and experimental results are given to illustrate the performance gain in using LCR method.

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

Access this chapter

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Amza, C., Cox, A. L., Warkadas, S. D., Kelehr, P., Lu, H., Rajamony, R., Yu, W., Zwaenepoel, W.: TreadMarks: Shared Memory Computing on Networks of Workstations. IEEE Computer, Vol. 29, 2(1996) 18–28.

    Google Scholar 

  2. Bolosky, W. J., Scott, M. L.: False Sharing and its Effect on Shared Memory Performance. Proc. of the USENIX Symposium on Experiences with Distributed and Multiprocessor Systems, (1993) 57–72

    Google Scholar 

  3. Chen, P. M., Lee, E. K., Gibson, G. A., Katz, R. H., Patterson, D. A.: RAID: High-Performance, Reliable Secondary Storage. ACM Computing Surveys, Vol. 26, 2(1994) 145–185.

    Article  Google Scholar 

  4. Gibson, G. A.: Redundant Disk Arrays: Reliable, Parallel Secondary Storage. MIT Press (1992).

    Google Scholar 

  5. Gibson, G. A., Nagle, D. F., Amiri, K., Chang, F. W., Feinberg, E. M. Gobioff, H., Lee, C., Ozceri, B., Riedel, E., Rochberg, D., Zelenka, J.: File Server, Scaling with Network-Attached Secure Disks. Proceedings of the ACM Int'l Conference on Measurement and Modeling of Computer Systems, (1997).

    Google Scholar 

  6. Hwang, K., Jin, H., Chow, E., Wang, C.-L., Xu, Z.: Designing SSI Clusters with Hierarchical Checkpointing and Single I/O Space. IEEE Concurrency, Vol. 7, 1(1999) 60–69

    Article  Google Scholar 

  7. Hwang, K., Xu, Z.: Scalable Parallel Computing: Technology, Architecture, programming. WCB/McGraw-Hill Co. (1998).

    Google Scholar 

  8. Jin, H., Hwang, K.: Case Studies of False Sharing Problems in the Cluster-based Disk Array. Proc. of 1999 ACM Symposium on Applied Computing, San Antonio, Texas (1999)

    Google Scholar 

  9. Li, K.: IVY: A Shared Virtual Memory System for Parallel Computing. Proc. of 1988 International Conference on Parallel Processing, Vol. II (1988), 94–101

    Google Scholar 

  10. Lee, E., Katz, R.: The Performance of Parity Placement in Disk Arrays. IEEE Trans. on Computers, Vol. C-42, 6(1993), 651–664.

    Article  Google Scholar 

  11. Menascé, D. A., Pentakalos, O. I., Yesha, Y.: An Analytic Model of Hierarchical Mass Storage Systems with Network-Attached Storage Devices. Proc. of the ACM International Conference on Measurement and Modeling of Computer Systems, (1996) 180–189.

    Google Scholar 

  12. Pfister, G.F.: In Search of Clusters. 2nd edn. Prentice-Hall, New Jersey (1998).

    Google Scholar 

  13. Quantum Co.: Storage Basics. http://www.quantum.com/src/storage_basics/

    Google Scholar 

  14. Stonebraker, M., Schloss, G. A.: Distributed RAID—a New Multiple Copy Algorithm. Proc. of the Sixth International Conference on Data Engineering, (1990) 430–443.

    Google Scholar 

  15. Torrellas, J., Lam, M. S., Hennessy, J. L.: False Sharing and Spatial Locality in Multiprocessor Caches. IEEE Trans. on Computers, Vol. C-43, 6(1994) 651–663.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Editor information

Constantine Polychronopoulos Kazuki Joe Akira Fukuda Shinji Tomita

Rights and permissions

Reprints and permissions

Copyright information

© 1999 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Jin, H., He, J., Chen, Q., Hwang, K. (1999). Grouped RAID accesses to reduce false sharing effect in clusters with single I/O space. In: Polychronopoulos, C., Fukuda, K.J.A., Tomita, S. (eds) High Performance Computing. ISHPC 1999. Lecture Notes in Computer Science, vol 1615. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0094934

Download citation

  • DOI: https://doi.org/10.1007/BFb0094934

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-65969-3

  • Online ISBN: 978-3-540-48821-7

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics