Comparing single-node and multi-node performance of an important fusion HPC code benchmark

  title={Comparing single-node and multi-node performance of an important fusion HPC code benchmark},
  author={Emily A. Belli and Jeff Candy and Igor Sfiligoi and Frank W{\"u}rthwein},
  journal={Practice and Experience in Advanced Research Computing},
Fusion simulations have traditionally required the use of leadership scale High Performance Computing (HPC) resources in order to produce advances in physics. The impressive improvements in compute and memory capacity of many-GPU compute nodes are now allowing for some problems that once required a multi-node setup to be also solvable on a single node. When possible, the increased interconnect bandwidth can result in order of magnitude higher science throughput, especially for communication… 

Figures and Tables from this paper

GX: a GPU-native gyrokinetic turbulence code for tokamak and stellarator design

GX is a code for solving the nonlinear gyrokinetic system for low-frequency turbulence in magnetized plasmas, particularly tokamaks and stellarators and is built to natively target graphics processors (GPUs), which are the fastest computational platforms available today.



Multiscale-optimized plasma turbulence simulation on petascale architectures

A high-accuracy Eulerian gyrokinetic solver for collisional plasmas

New Compute Engine A2 VMs—first NVIDIA Ampere A100 GPUs in the cloud

  • Google Blog
  • 2020


  • Retrieved March 31
  • 2022


  • Retrieved March 31
  • 2022

NVIDIA Technical White Paper

  • 2020

CGYRO. Computer Software. USDOE Office of Science (SC), Fusion Energy Sciences (FES) (SC-24)

  • 2018

Summit User Guide

  • Retrieved March 31
  • 2022