Automated performance modeling based on runtime feature detection and machine learning

Jingwei Sun, Shiyan Zhan, Guangzhong Sun, Yong Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Automated performance modeling and performance prediction of parallel programs are highly valuable in many use cases, such as in guiding task management and job scheduling, offering insights of application behaviors, assisting resource requirement estimation, etc. The performance of parallel programs is affected by numerous factors, including but not limited to hardware, system software, applications, algorithms, and input parameters, thus an accurate performance prediction is often a challenging and daunting task. In this study, we focus on automatically predicting the execution time of parallel programs (more specifically, MPI programs) with different inputs, at different scale, and without domain knowledge. We model the correlation between the execution time and domain-independent runtime features. These features include values of variables, counters of branches, loops, and MPI communications. Through automatically instrumenting an MPI program, each execution of the program will output a feature vector and its corresponding execution time. After collecting data from executions with different inputs, a random forest machine learning approach is used to build an empirical performance model, which can predict the execution time of the program with a new input. Our experiments and analyses of three parallel programs, Graph500, GalaxSee and SMG2000, on three different systems show that our method performs well, with less than 20% error in predictions on average.

Original languageEnglish
Title of host publicationProceedings - 15th IEEE International Symposium on Parallel and Distributed Processing with Applications and 16th IEEE International Conference on Ubiquitous Computing and Communications, ISPA/IUCC 2017
EditorsGregorio Martinez, Richard Hill, Geoffrey Fox, Peter Mueller, Guojun Wang
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages744-751
Number of pages8
ISBN (Electronic)9781538637906
DOIs
StatePublished - May 25 2018
Event15th IEEE International Symposium on Parallel and Distributed Processing with Applications and 16th IEEE International Conference on Ubiquitous Computing and Communications, ISPA/IUCC 2017 - Guangzhou, China
Duration: Dec 12 2017Dec 15 2017

Publication series

NameProceedings - 15th IEEE International Symposium on Parallel and Distributed Processing with Applications and 16th IEEE International Conference on Ubiquitous Computing and Communications, ISPA/IUCC 2017

Conference

Conference15th IEEE International Symposium on Parallel and Distributed Processing with Applications and 16th IEEE International Conference on Ubiquitous Computing and Communications, ISPA/IUCC 2017
CountryChina
CityGuangzhou
Period12/12/1712/15/17

Keywords

  • Machine learning
  • Parallel computing
  • Performance modeling

Fingerprint Dive into the research topics of 'Automated performance modeling based on runtime feature detection and machine learning'. Together they form a unique fingerprint.

  • Cite this

    Sun, J., Zhan, S., Sun, G., & Chen, Y. (2018). Automated performance modeling based on runtime feature detection and machine learning. In G. Martinez, R. Hill, G. Fox, P. Mueller, & G. Wang (Eds.), Proceedings - 15th IEEE International Symposium on Parallel and Distributed Processing with Applications and 16th IEEE International Conference on Ubiquitous Computing and Communications, ISPA/IUCC 2017 (pp. 744-751). (Proceedings - 15th IEEE International Symposium on Parallel and Distributed Processing with Applications and 16th IEEE International Conference on Ubiquitous Computing and Communications, ISPA/IUCC 2017). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ISPA/IUCC.2017.00115