Harjinder S. Sandhu
Abstract
The increasing need for data sharing in large-scale distributed systems may place a heavy burden on critical resources such as file servers and networks. Our examination of the workload in one large commercial engineering environment shows that wide-spread sharing of unstable files among tens to hundreds of users is common. Traditional client-based file cacheing techniques are not scalable in such environments.
We propose Frolic, a scheme for cluster-based file replication in large-scale distributed file systems. A cluster is a group of workstations and one or more file servers on a local area network. Large distributed systems may have tens or hundreds of clusters connected by a backbone network. By dynamically creating and maintaining replicas of shared files on the file servers in the clusters using those files, we effectively reduce reliance on central servers supporting such files, as well as reduce the distances between the accessing sites and data. We propose and study algorithms for the two main issues in Frolic, 1) locating a valid file replica, and 2) maintaining consistency among replicas. Our simulation experiments using a statistical workload model based upon measurement data and real workload characteristics show that cluster-based file replication can significantly reduce file access delays and server and backbone network utilizations in large-scale distributed systems over a wide range of workload conditions. The workload characteristics most critical to replication performance are: the size of shared files, the number of clusters that modify a file, and the number of consecutive accesses to files from a particular cluster.
- BHK+91 M. Baker, J. Hartman, M. Kupfer, K. Shirriff, and J. Ousterhout. Measurements of a distributed file system. In Proc. 13th A CM Symposium on Operating System Principles. ACM, October 1991. Google Scholar
Digital Library
- BR90 P. Biswas and K.K. Ramakrishnan. File access characterization of VAX/VMS environments. In l Oth Int. Conf. on Distributed Computing Systems, November 1990.Google Scholar
- HKM+88 John Howard, M. Kazar, S. Menees, D. Nichols, M. Satyanarayanan, R. Sidebotham, and M. West. Scale and performance in a distributed file system. A CM Transactions on Computer Systems, 6(1), February 1988. Google ScholarDigital Library
- Kur88 Q. Kure. Optimization of File Migration in Distributed Systems. PhD thesis, University of California Berkeley, 1988. Google ScholarDigital Library
- Li88 K. Li. Memory coherence in shared virtual memory systems. A CM Transactions on Computer Systems, Vol 7, No 4, November 1989. Google ScholarDigital Library
- Mol90 M. Molloy. Validation of MVA models for client/server systems. Hewlett-Packard, 1990.Google Scholar
- OCH+85 J. Ousterhout, H. Da Costa, D. Harrison, J. Kunze, M. Kupfer, and J. Thompson. A trace-driven analysis of the Unix 4.2 BSD file system. In Proc. l Oth A CM Symposium on Operating System Principles. ACM, December 1985. Google ScholarDigital Library
- San91 H.S. Sandhu. File replication and performance in large-scale distributed systems. Master's thesis, University of Toronto, January 1991.Google Scholar
- Sch88 H. Schwetman. Using CSIM to model complex systems. In Proceeding of the 1988 Winter Simulation Conference, 1988. Google ScholarDigital Library
- SGK+85 R. Sandberg, D. Goldberg, S. Kleiman, D. Walsh, and B. Lyon. Design and implementation of the Sun network file system. In Useuiz Conference and E~hibition, Portland, OR~ Summer 1985.Google Scholar
- SZ90 M. Stumm and S. Zhou. Algorithms implementing distributed shared memory. Computer, May 1990. Google ScholarDigital Library
- SZ92 H.S. Sandhu and S. Zhou. A case study of file system workload in a large-scale distributed environment. Technical report, University of Toronto, 1992. In preparation.Google Scholar
- ZS90 S. Zhou and C. Siebenmann. Snooper user guide, January 1990. University of Toronto.Google Scholar
Index Terms
- Cluster-based file replication in large-scale distributed systems
Recommendations
Cluster-based file replication in large-scale distributed systems
SIGMETRICS '92/PERFORMANCE '92: Proceedings of the 1992 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systemsThe increasing need for data sharing in large-scale distributed systems may place a heavy burden on critical resources such as file servers and networks. Our examination of the workload in one large commercial engineering environment shows that wide-...
Comments