At SC20: A bridge to ‘Ponte Vecchio’: Argonne Aurora builders utilizing substitute Intel Xe HP GPUs
Intel and Argonne National Laboratory announced today that they are using GPUs based on Intel’s Xe-HP microarchitecture and Intel oneAPI toolkits to develop scientific applications for the Aurora exascale system, pending future delivery of Intel 7nm Ponte Vecchio GPUs. This will power Aurora when the delayed system is deployed in 2022.
Objective: To use Intel’s heterogeneous computer programming environments so that scientific applications are ready when deployed for the size and architecture of the Aurora supercomputer.
The Aurora system (A21), for which Intel is the prime contractor, was scheduled for use in 2021 and would be the first exascale-class system in the United States. However, in July, Intel announced that its ‘Ponte Vecchio’ GPU, a key Aurora component, would be delayed by at least six months, meaning HPE-Crays Frontier, which runs on AMD CPUs and GPUs, and for the Oak Ridge National Laboratory is destined to be the first exascale system in the country. (Note that in its HPC Market Update yesterday, industry analyst firm Hyperion Research expected Aurora to ship about 12 months behind schedule.)
In its announcement today, Intel said that researchers at the Argonne Leadership Computing Facility (ALCF) are using software development platforms based on Intel Xe-HP GPUs as part of the Aurora Early Science Program and Exascale Computing Project that enable critical applications and Libraries should be prepared and infrastructure for Aurora. Intel and Argonne teams work together to jointly design, test, and validate multiple exascale applications.
“With access to Intel Xe HP GPUs and development tools, Argonne developers can perform software optimizations for Intel CPUs and GPUs and examine scenarios that are difficult to replicate in software-only environments,” said Intel. “The Xe HP GPUs provide Intel Xe HPC GPUs (Ponte Vecchio) with a development vehicle that will be used in the Aurora system.”
“Our collaboration with Intel in developing code between architectures benefits many of our development teams,” said Susan Coghlan, project leader for Aurora at ALCF. “This co-design approach has resulted in the software stack rapidly maturing to production quality for execution on Aurora.”
Examples of development work are:
- The EXAALT project, which enables molecular dynamics in the exascale for fusion and fission energy problems.
- The QMCPACK project developing Exascale Quantum Monte Carlo algorithms to improve predictions about complex materials.
- The GAMESS project, which develops ab initio fragmentation methods to address challenges in computational chemistry such as heterogeneous catalysis problems more efficiently.
- The ExaSMR project, which develops high-fidelity modeling functions at exascale for complex physical phenomena that occur in the operation of nuclear reactors in order to ultimately improve their design.
- The HACC project, which develops on exascal extreme-scale cosmological simulations that allow scientists to simultaneously analyze observational data from state-of-the-art telescopes to test various theories.