Home Publications edited volumes Awards Research Teaching Miscellaneous Full CV [pdf] BLOG
Events
Past Events

Publications of Torsten Hoefler
Marcus Ritter, Alexander Geiss, Johannes Wehrstein, Alexandru Calotoiu, Thorsten Reimann, Torsten Hoefler, Felix Wolf:
  NoiseResilient Empirical Performance Modeling with Deep Neural Networks
(In IPDPS '21: Proceedings of the 35th IEEE Interational Parallel and Distributed Processing Symposium, May 2021)
AbstractEmpirical performance modeling is a proven instrument to analyze the scaling behavior of HPC applications.
Using a set of smallerscale experiments, it can provide important
insights into application behavior at larger scales. ExtraP is an
empirical modeling tool that applies linear regression to automatically generate humanreadable performance models. Similar
to other regressionbased modeling techniques, the accuracy of
the models created by ExtraP decreases as the amount of noise
in the underlying data increases. This is why the performance
variability observed in many contemporary systems can become
a serious challenge. In this paper, we introduce a novel adaptive
modeling approach that makes ExtraP more noise resilient,
exploiting the ability of deep neural networks to discover the
effects of numerical parameters, such as the number of processes
or the problem size, on performance when dealing with noisy
measurements. Using synthetic analysis and data from three
different case studies, we demonstrate that our solution improves
the model accuracy at high noise levels by up to 25% while
increasing their predictive power by about 15%.
Documentsdownload article:
  BibTeX  @inproceedings{, author={Marcus Ritter and Alexander Geiss and Johannes Wehrstein and Alexandru Calotoiu and Thorsten Reimann and Torsten Hoefler and Felix Wolf}, title={{NoiseResilient Empirical Performance Modeling with Deep Neural Networks}}, year={2021}, month={May}, booktitle={IPDPS '21: Proceedings of the 35th IEEE Interational Parallel and Distributed Processing Symposium}, source={http://www.unixer.de/~htor/publications/}, } 

