A Content-based Load Balancing Algorithm for Metadata Servers in Cluster File System

클러스터 파일 시스템의 메타데이터 서버를 위한 내용 기반 부하 분산 알고리즘

  • Published : 2006.08.01


A metadata service is one of the important factors to affect the performance of cluster file systems. We propose a content-based load balancing algorithm that dynamically distributes client requests to appropriate metadata servers based on the types of metadata operations. By replicating metadatas and logging update messages in each server, rather than moving metadatas across servers, we significantly reduced the response time and evenly distributed client's requests among metadata servers.


  1. SPEC, 'SFS 3.0 Documentation Version 1.0,' Standard Performance Evaluation Corporation, 2001
  2. K. W. Preslan et al., 'A 64 Bit, Shared Disk File System for Linux,' Proceedings of the 16th IEEE Mass Storage Systems Symposium, pp.22-41, 1999
  4. P. H. Carns et al, 'PVFS: A Parallel File System For Linux Clusters,' Proceedings of the 4th Annual Linux Showcase and Conference, Atlanta, pp.317-327, Oct., 2000
  5. Peter J. Braam et al., 'The Lustre Storage Architecture,' Cluster File System. Inc, Mar., 2003
  6. Scott A. Brandt et al., 'Efficient Metadata Management in Large Distributed Storage Systems,' Proceedings of the 11th IEEE NASA Goddard Conference on Mass Storage Systems and Technologies, Apr., 2003
  7. Jin Xiong et al., 'Design and Perforamance of the the Dawning Cluster File System,' IEEE International Conference on Cluster Computing (Cluster'03), Dec., 2003
  8. Peter F. Corbett et al., 'The Vesta parallel file system,' ACM Transactions on Computer Systems(TOCS), Vol.14, No.3, pp.225-264, Aug., 1996
  9. Daniel P. Bovet et al, Understanding the Linux Kernel, O'Reilly and Associates, Sebastopol, 2003
  11. L. mummert and M. Satyanarayanan. 'Long term distributed file reference tracing: Implementation and experience'. Software-Practice and Experience (SPE), Vol.26, No.6, pp.705-736, June, 1996<705::AID-SPE29>3.3.CO;2-F
  12. Kishor Shridharbhai Trivedi, Probability and Statistics with Reliability, Queuing, and Computer Science Applications, John Wiley & Sons, Inc., New York, 2002