Identifying code phases using piece-wise linear regressions
Document typeConference report
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Rights accessRestricted access - publisher's policy
Node-level performance is one of the factors that may limit applications from reaching the supercomputers' peak performance. Studying node-level performance and attributing it to the source code results into valuable insight that can be used to improve the application efficiency, albeit performing such a study may be an intimidating task due to the complexity and size of the applications. We present in this paper a mechanism that takes advantage of combining piece-wise linear regressions, coarse-grain sampling, and minimal instrumentation to detect performance phases in the computation regions even if their granularity is very fine. This mechanism then maps the performance of each phase into the application syntactical structure displaying a correlation between performance and source code. We introduce a methodology on top of this mechanism to describe the node-level performance of parallel applications, even for first-time seen applications. Finally, we demonstrate the methodology describing optimized in-production applications and further improving their performance applying small transformations to the code based on the hints discovered. © 2014 IEEE.
CitationServat, H. [et al.]. Identifying code phases using piece-wise linear regressions. A: IEEE International Parallel and Distributed Processing Symposium. "IEEE 28th International Parallel and Distributed Processing Symposium (IPDPS 2014): proceedings: Phoenix, Arizona, USA: 19-23 May 2014". Phoenix: Institute of Electrical and Electronics Engineers (IEEE), 2014, p. 941-951.
|Identifying cod ... ise linear regressions.pdf||Identifying code phases using piece-wise linear regressions.pdf||1.947Mb||Restricted access|