Comparison of Graphcore IPUs and Nvidia GPUsfor cosmology applications

2021/06/04 by Bastien Arcelin, Arcelin, Bastien
Computer Science · Engineering · Physics and Astronomy · #CCD and CMOS Imaging Sensors #Computational Physics (physics.comp-ph) #FOS: Physical sciences #Galaxies: Formation, Evolution, Phenomena #Gaussian Processes and Bayesian Inference #Instrumentation and Methods for Astrophysics (astro-ph.IM)

paper · pdf · doi:10.48550/arxiv.2106.02465

openalex publication_date 2021/06/04 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

This paper represents the first investigation of the suitability and performance of Graphcore Intelligence Processing Units (IPUs) for deep learning applications in cosmology. It presents the benchmark between a Nvidia V100 GPU and a Graphcore MK1 (GC2) IPU on three cosmological use cases: a classical deep neural network and a Bayesian neural network (BNN) for galaxy shape estimation, and a generative network for galaxy images production. The results suggest that IPUs could be a potential avenue to address the increasing computation needs in cosmology.

Citations

• Adam: A Method for Stochastic Optimization
• Generative Adversarial Networks
• Auto-Encoding Variational Bayes
• Wide-Field InfrarRed Survey Telescope-Astrophysics Focused Telescope Assets WFIRST-AFTA 2015 Report
• Weight Uncertainty in Neural Networks
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Dissecting the Graphcore IPU Architecture via Microbenchmarking
• fink, a new generation of broker for the LSST community

Related

• Studying the potential of Graphcore IPUs for applications in Particle Physics
• Application of GPUs for the Calculation of Two Point Correlation\n Functions in Cosmology
• Fitting Galaxies on GPUs
• Advanced Architectures for Astrophysical Supercomputing
• Graphcore C2 Card performance for image-based deep learning application: A Report