Scalable Relative Debugging

Abstract

Detecting and isolating bugs that arise only at high processor counts is a challenging task. Over a number of years, we have implemented a special debugging method, called "relative debugging," that supports debugging applications as they evolve or are ported to larger machines. It allows a user to compare the state of a suspect program against another reference version even as the number of processors is increased. The innovative idea is the comparison of runtime data to reason about the state of the suspect program. While powerful, a naïve implementation of the comparison phase does not scale to large problems running on large machines. In this paper, we propose two different solutions including a hash-based scheme and a direct point-to-point scheme. We demonstrate the implementation, a case study, as well as the performance, of our techniques on 20K cores of a Cray XE6 system.Index Terms-Parallellism and concurrency, distributed debugging, assertion checkers Ç . M.N. Dinh is with the Faculty

Augmenting Diffs With Runtime Information

“…Abramson et al have introduced relative debugging [62], [63], [64], [65], [66], [67]. Relative debugging is implemented in their tool, called GUARD.…”

A Survey of Algorithmic Debugging

“…As outlined before, the recorded debugging sessions can be considered as repositories of test cases that can be used to automatically test the evolved software. This idea is very related to another approach that also records and profits the previous debugging sessions: relative debugging [Abramson et al 1996], a technique that has been successfully applied to several environments [Rose et al 2015;Dinh et al 2014;Sosic and Abramson 1997]. Relative debugging frameworks allow the user to establish a comparison between the execution of two programs.…”