Advertisement

An Implementation of Semi-synchronous Remote Mirroring System for SANs

  • Rui Yan
  • Ji-wu Shu
  • Dong-chan Wen
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3252)

Abstract

Remote mirroring is often used as part of disaster recovery solutions. Synchronous remote mirroring incurs steep costs in both write latency and network bandwidth to support the mirroring, while asynchronous mirroring does not ensure the consistency of remote data. In this paper, we designed and implemented a storage-based semi-synchronous remote mirroring system for SANs. By using a log policy for the active remote write commands, this approach allows a limited number of write I/O operations to proceed before waiting for acknowledgment of receipt from the remote site, which significantly reduces write latency. This implementation also provides a consistent copy in a remote site to meet the demand for disaster recovery, because the order of commands is guaranteed. Furthermore, it can be applied to the working condition of long distance mirroring with high network latency. The testing results show that with the same request size and network latency, our semi-synchronous remote mirroring reduces average write command response time by 14-20% compared with synchronous remote mirroring.

Keywords

Remote Site Average Response Time Network Latency Storage Node Disaster Recovery 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Azagury, A.C., Factor, M.E., Micka, W.F.: Advanced Functions for Storage Subsystems: Supporting Continuous Availability. IBM SYSTEM Journal 42 (2003)Google Scholar
  2. 2.
    Using EMC SnapView and MirrorView for Remote Backup, Engineering White Paper, EMC Corporation (April 2002) Google Scholar
  3. 3.
    Software Solutions Guide for Enterprise Storage, Hitachi Data Systems Corporation (December 2000) Google Scholar
  4. 4.
    Patterson, R.H., et al.: SnapMirror: File-System-Based Asynchronous Mirroring for Disaster Recovery. In: First USENIX conference on File and Storage Technologies, Monterey, CA, USA (2002)Google Scholar
  5. 5.
    Ervin, N.E.: Managing Extended Distance EMC SRDF Semi-Synchronous Remote Copy. Technical report (August 1999)Google Scholar
  6. 6.
    Ji-wu, S., et al.: A Highly Efficient FC-SAN Based On Load Stream. In: Zhou, X., Xu, M., Jähnichen, S., Cao, J. (eds.) APPT 2003. LNCS, vol. 2834, pp. 31–40. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  7. 7.
    Technical Report: Design and Implementation of the TH-MSNS (In Chinese), Computer Science Department, Tsinghua University, P.R. China (2003)Google Scholar
  8. 8.
    Palekar, A., Ganapathy, N.: Design and Implement of A LINUX SCSI Target for Storage Area Networks. In: Proceedings of the 5th Annual Linux Showcase & Conference, Oakland, USA (2001)Google Scholar
  9. 9.
    Namgoong, J.-C., Park, C.-I.: Design and Implement of a Fibre Channel Network Driver for SAN-Attached RAID Controllers. IEEE Parallel and Distributed Systems (2001)Google Scholar
  10. 10.
    Ji-wu, S., Rui, Y., Chan, W.D., Wei-min, Z.: An Implementation of Storage-Based Synchronous Remote Mirroring for SANs. In: IFIP International Conference on Network and Parallel Computing (2004) (accepted)Google Scholar
  11. 11.
    Jun, Y., Ji-wu, S., Wei-min, Z.: A Distributed Storage Cluster Design for Remote Mirroring based on Storage Area Network. IEEE Transactions on computers (submitted), also in http://storage.cs.tsinghua.edu.cn
  12. 12.
    Sievert, J.: Iometer: The I/O Performance Analysis Tool for Servers, http://www.intel.com/design/servers/devtools/iometer/index.htm

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Rui Yan
    • 1
  • Ji-wu Shu
    • 1
  • Dong-chan Wen
    • 1
  1. 1.Department of Computer Science and TechnologyTsinghua UniversityBeijingChina

Personalised recommendations