Performance evaluations of a parallel, scalable and expandable database computer

Looks at performance evaluations of a experimental database computer with a variable number of parallel database backends. It focuses on (1) the benchmarking methodology for parallel database computers, (2) architectural significance of a parallel database computer, and (3) the benchmark results and interpretations of the computer´s performance. Because of its strong influence on methodology, the paper begins with a brief introduction to the architecture of the database computer. The kinds of parallelism, scalability and expandability are articulated. The establishment of benchmarking methodology and performance measures employed follows. Finally, the results of benchmarks and their interpretations are presented. The paper shows innovative benchmarking methodology and performance measures for parallel database computers and indicates their viability and promise for performance gains and capacity growth