Duet Benchmarking: Improving Measurement Accuracy in the Cloud

Abstract

We investigate the duet measurement procedure, which helps improve the accuracy of performance comparison experiments conducted on shared machines by executing the measured artifacts in parallel and evaluating their relative performance together, rather than individually. Specifically, we analyze the behavior of the procedure in multiple cloud environments and use experimental evidence to answer multiple research questions concerning the assumption underlying the procedure. We demonstrate improvements in accuracy ranging from 2.3× to 12.5× (5.03× on average) for the tested ScalaBench (and DaCapo) workloads, and from 23.8× to 82.4× (37.4× on average) for the SPEC CPU 2017 workloads.