HOLMES: Effective statistical debugging via efficient path profiling

TLDR

Statistical debugging automates bug isolation by profiling multiple program runs and using statistical analysis to pinpoint likely failure causes. The study investigates how richer path profiles and an adaptive HOLMES tool can improve bug isolation effectiveness. HOLMES isolates bugs by identifying failure‑correlated paths, and its adaptive variant iteratively profiles bugs to reduce execution time and space, evaluated on SIR benchmarks and real‑world programs. Path profiles enable more precise bug isolation by revealing contextual information, and bug‑directed profiling achieves scalable, accurate isolation with low overhead.

Abstract

Statistical debugging aims to automate the process of isolating bugs by profiling several runs of the program and using statistical analysis to pinpoint the likely causes of failure. In this paper, we investigate the impact of using richer program profiles such as path profiles on the effectiveness of bug isolation. We describe a statistical debugging tool called HOLMES that isolates bugs by finding paths that correlate with failure. We also present an adaptive version of HOLMES that uses iterative, bug-directed profiling to lower execution time and space overheads. We evaluate HOLMES using programs from the SIR benchmark suite and some large, real-world applications. Our results indicate that path profiles can help isolate bugs more precisely by providing more information about the context in which bugs occur. Moreover, bug-directed profiling can efficiently isolate bugs with low overheads, providing a scalable and accurate alternative to sparse random sampling.