Data-flow query processing using multiprocessor hash-partitioned algorithms

Thesis/Dissertation · Wed Jan 01 04:00:00 EST 1986

OSTI ID:5254628

Gerber, R H

This thesis demonstrates that hash-partitioned query-processing algorithms can serve as a basis for a highly parallel, high-performance relational data-base machine. In addition to demonstrating that parallelism can really be made to work in a data-base machine context, it is shown that such parallelism can be controlled with minimal overhead using data-flow query-processing techniques that pipeline data between highly autonomous, distributed processes. For this purpose, the design, implementation techniques, and initial performance evaluation of Gamma, a new relational data-base machine are presented. Gamma is a fully operational prototype consisting of 20 VAX 11/750 computers. The Gamma architecture illustrates that a high-performance data-base machine can be constructed without the assistance of special-purpose hardware components. Finally, a simulation model of Gamma is presented that accurately reflects the measured performance of the actual Gamma prototype. Using this simulation model, the performance of Gamma is explored for large multiprocessor systems with varying hardware capabilities.

• Research Organization: Wisconsin Univ., Madison (USA)
• OSTI ID: 5254628
• Country of Publication: United States
• Language: English

Related Subjects

99 GENERAL AND MISCELLANEOUS
990210* -- Supercomputers-- (1987-1989)
990300 -- Information Handling
ALGORITHMS
ARRAY PROCESSORS
DATA PROCESSING
DATA-FLOW PROCESSING
DISTRIBUTED DATA PROCESSING
MATHEMATICAL LOGIC
PARALLEL PROCESSING
PROCESSING
PROGRAMMING
SIMULATION