Supporting partial data accesses to replicated data

Partial data access operations occur frequently in distributed systems. This paper presents new approaches for efficiently supporting partial data access operations to replicated data. We propose the replica modularization (RM) technique which suggests partitioning replicas into modules, which now become the minimum unit of data access. RM is shown to increase the availability of both partial read and write operations and improves performance by reducing access delays and the size of data transfers occurring during operation execution on replicated data. In addition, we develop a new module-based protocol (MB) in which different replication protocols are used to access different sets of replicas, with each replica storing different modules. The instance of MB we discuss here is a hybrid of the ROWA (Read One Write All) protocol and the MQ (Majority Quorum) protocol. MB allows a trade-off between storage costs and availability. We show that MB can achieve almost as high availability as the MQ protocol, but with considerably smaller storage costs.<<ETX>>

[1]  Divyakant Agrawal,et al.  The generalized tree quorum protocol: an efficient approach for managing replicated data , 1992, TODS.

[2]  P. Triantafillou,et al.  High availability is not enough (distributed systems) , 1992, [1992 Proceedings] Second Workshop on the Management of Replicated Data.

[3]  Edward D. Lazowska,et al.  Improving fault tolerance and supporting partial writes in structured coterie protocols for replicated objects , 1992, SIGMOD '92.

[4]  Akhil Kumar,et al.  Hierarchical Quorum Consensus: A New Algorithm for Managing Replicated Data , 1991, IEEE Trans. Computers.

[5]  Peter Triantafillou,et al.  Using multiple replica classes to improve performance in distributed systems , 1991, [1991] Proceedings. 11th International Conference on Distributed Computing Systems.

[6]  Peter Triantafillou,et al.  Efficiently maintaining availability in the presence of partitionings in distributed systems , 1991, [1991] Proceedings. Seventh International Conference on Data Engineering.

[7]  Peter Triantafillou,et al.  A new paradigm for high availability and efficiency in replicated distributed databases , 1990, Proceedings of the Second IEEE Symposium on Parallel and Distributed Processing 1990.

[8]  Divyakant Agrawal,et al.  The Tree Quorum Protocol: An Efficient Approach for Managing Replicated Data , 1990, VLDB.

[9]  Sushil Jajodia,et al.  Dynamic voting algorithms for maintaining the consistency of a replicated database , 1990, TODS.

[10]  Akhil Kumar,et al.  Performance analysis of a hierarchical quorum consensus algorithm for replicated objects , 1990, Proceedings.,10th International Conference on Distributed Computing Systems.

[11]  Mostafa H. Ammar,et al.  The grid protocol: a high performance scheme for maintaining replicated data , 1990, [1990] Proceedings. Sixth International Conference on Data Engineering.

[12]  Divyakant Agrawal,et al.  Reducing Storage for Quorum Consensus Algorithms , 1988, VLDB.

[13]  Sushil Jajodia,et al.  Dynamic voting , 1987, SIGMOD '87.

[14]  Philip A. Bernstein,et al.  Concurrency Control in Distributed Database Systems , 1986, CSUR.

[15]  David K. Gifford,et al.  Weighted voting for replicated data , 1979, SOSP '79.

[16]  Peter Triantafillou,et al.  Recovering in large distributed systems with replicated data , 1993, [1993] Proceedings of the Second International Conference on Parallel and Distributed Information Systems.

[17]  Darrell D. E. Long,et al.  The management of replication in a distributed system , 1988 .