ICL Research Profile

HPCG

Overview

The High Performance Conjugate Gradients (HPCG) benchmark is designed to measure performance that is representative of modern scientific applications. It does so by exercising the computational and communication patterns commonly found in real science and engineering codes, which are often based on sparse iterative solvers. HPCG exhibits the same irregular accesses to memory and fine-grain recursive computations that dominate large-scale scientific workloads used to simulate complex physical phenomena.

The HPCG 3.0 reference code was released on November 11, 2015 for the SC15 conference in Austin, TX. In addition to bug fixes, this release positions HPCG to even better represent modern partial differential equation (PDE) solvers and aids in running HPCG on production supercomputing installations. The reference version is accompanied by binary releases from Intel and NVIDIA that are carefully optimized for the vendors’ respective hardware platforms. The current HPCG performance list was released at SC17 and now features over 100 supercomputing sites. The HPCG score has also been tracked by the TOP500 list since June 2017.

Find out more at http://www.hpcg-benchmark.org

In Collaboration With

  1. Sandia National Laboratories

Sponsored by

  1. National Nuclear Security Administration
  2. The United States Department of Energy

Project Handout