A performance comparison of two architectures for fast transaction processing

Abstract

Investigates the issues involved in using multiprocessors for transaction processing. The authors use a simulation model to study the behavior of two different architectures, namely shared everything and shared nothing. In shared everything, any processor can access any disk and all memory is shared. In shared nothing, neither disks nor memory is shared. They study the effects of data contention and resource contention in both of these architectures. They quantify the effects of intraquery parallelism on both architectures under different operating conditions.<>