Capturing and automating performance diagnosis: the Poirot approach

Petiormance diagnosis, the process of finding and ex-plaining performance problems, is an important part of parallel programming. Effective pelformunce diagnosis requires that the programmer plan an appropriate method, and manage the experiments required by that method. This paper presents Poirot, an architecture to support per&or-mrtnce diagnosis. It explains how the architecture helps au-tomatically, adaptably plan and manage the diagnosis pro-ce.rs. The paper evaluates the generality and practicality of Poirot, by reconstructing diagnosis methods.found in sev-eral published performance tools