Analysing an SQL application with a BSPlib call-graph profiling tool

Jonathan Hill and Stephen Jarvis and Constantinos Siniolakis and Vasil Vasilev

Abstract
This paper illustrates the use of a post-mortem call-graphprofiling tool in the analysis of an SQL query processingapplication written using BSPlib. Unlike other parallelprofiling tools, the architecture independent metric of imbalance in size of communicated data is used to guide program optimisation. We show that by using this metric, BSPlib programs can be optimised in a portable and architecture independent manner. Results are presented to support this claim for unoptimised and optimised versions of a programrunning on networks of workstations, shared memory multiprocessors and tightly coupled distributed memory parallel machines.
Contact
Stephen Jarvis
Dr. Stephen Jarvis,,Oxford University Computing Laboratory,,Wolfson Building,,Parks Road,,Oxford. ,OX1 3QD
Stephen.Jarvis@comlab.ox.ac.uk