首页|Reproducibility of the DaCe Framework on NPBench Benchmarks

Reproducibility of the DaCe Framework on NPBench Benchmarks

扫码查看
原文链接
  • NETL
  • NSTL
  • IEEE
DaCe is a framework for Python that claims to provide massive speedups with C-like speeds compared to already existing high-performance Python frameworks (e.g. Numba or Pythran). In this work, we take a closer look at reproducing the NPBench work. We use performance results to confirm that NPBench achieves higher performance than NumPy in a variety of benchmarks and provide reasons as to why DaCe is not truly as portable as it claims to be, but with a small adjustment it can run anywhere.

Benchmark testingGraphics processing unitsPythonRandom access memoryHardwareOptimizationSoftware

Anish Govind、Yuchen Jing、Stefanie Dao、Michael Granado、Rachel Handran、Davit Margarian、Matthew Mikhailov、Danny Vo、Matei-Alexandru Gardus、Khai Vu、Derek Bouius、Bryan Chin、Mahidhar Tatineni、Mary Thomas

展开 >

Electrical, Computer Engineering, University of California San Diego, La Jolla, CA, USA

Computer Science Engineering, University of California San Diego, La Jolla, CA, USA

Cognitive Science, University of California San Diego, La Jolla, CA, USA

Mathematics, University of California San Diego, La Jolla, CA, USA

Advanced Micro Devices, Santa Clara, CA, USA

San Diego Supercomputer Center, University of California San Diego, La Jolla, CA, USA

展开 >

2025

10.1109/TPDS.2024.3427130

IEEE transactions on parallel and distributed systems

IEEE transactions on parallel and distributed systems

SCI
ISSN：
年,卷(期)：2025.36(5)
  • 3