ytopt: Autotuning Scientific Applications for Energy Efficiency at Large Scales

Xingfu Wu, Prasanna Balaprakash, Michael Kruse, Jaehoon Koo, Brice Videau, Paul Hovland, Valerie Taylor, Brad Geltz, Siddhartha Jana, Mary Hall
Argonne National Laboratory, Lemont, IL 60439
arXiv:2303.16245 [cs.DC], (28 Mar 2023)


   title={ytopt: Autotuning Scientific Applications for Energy Efficiency at Large Scales},

   author={Xingfu Wu and Prasanna Balaprakash and Michael Kruse and Jaehoon Koo and Brice Videau and Paul Hovland and Valerie Taylor and Brad Geltz and Siddhartha Jana and Mary Hall},






As we enter the exascale computing era, efficiently utilizing power and optimizing the performance of scientific applications under power and energy constraints has become critical and challenging. We propose a low-overhead autotuning framework to autotune performance and energy for various hybrid MPI/OpenMP scientific applications at large scales and to explore the tradeoffs between application runtime and power/energy for energy efficient application execution, then use this framework to autotune four ECP proxy applications — XSBench, AMG, SWFFT, and SW4lite. Our approach uses Bayesian optimization with a Random Forest surrogate model to effectively search parameter spaces with up to 6 million different configurations on two large-scale production systems, Theta at Argonne National Laboratory and Summit at Oak Ridge National Laboratory. The experimental results show that our autotuning framework at large scales has low overhead and achieves good scalability. Using the proposed autotuning framework to identify the best configurations, we achieve up to 91.59% performance improvement, up to 21.2% energy savings, and up to 37.84% EDP improvement on up to 4,096 nodes.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2024 hgpu.org

All rights belong to the respective authors

Contact us: