Much ado about shared-nothing

In a 'shared-nothing' parallel computer, each processor has its own memory and disks and processors communicate by passing messages through an interconnect. Many academic researchers, and some vendors, assert that shared-nothingness is the 'consensus' architecture for parallel DBMSs. This alleged consensus is used as a justification for simulation models, algorithms, research prototypes and even marketing campaigns. We argue that shared-nothingness is no longer the consensus hardware architecture and that hardware resource sharing is a poor basis for categorising parallel DBMS software architectures if one wishes to compare the performance characteristics of parallel DBMS products.

[1]  Felipe Cariño,et al.  NCR 3700 - The Next-Generation Industrial Database Computer , 1993, VLDB.

[2]  Kien A. Hua,et al.  Interconnecting shared-everything systems for efficient parallel query processing , 1991, [1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems.

[3]  David J. DeWitt,et al.  Parallel database systems: the future of high performance database systems , 1992, CACM.

[4]  Patrick Valduriez,et al.  Parallel database systems: The case for shared-something , 1993, Proceedings of IEEE 9th International Conference on Data Engineering.

[5]  Michael Stonebraker,et al.  The Case for Shared Nothing , 1985, HPTS.

[6]  Michael Stonebraker,et al.  A performance comparison of two architectures for fast transaction processing , 1988, Proceedings. Fourth International Conference on Data Engineering.

[7]  Patrick Valduriez,et al.  Overview of Parallel Architectures for Databases , 1993, Comput. J..

[8]  Chaitanya K. Baru,et al.  DB2 Parallel Edition , 1995, IBM Syst. J..