ALCC program awards ALCF computing time to 16 projects
DOE's ASCR Leadership Computing Challenge has selected a new set of projects for 2021-2022.
In a new video from DOE’s Exascale Computing Project, ALCF’s Scott Parker provides an overview of how the ECP Application Integration effort is helping to ensure scientific applications are ready for the nation’s upcoming exascale supercomputers.
The Exascale Computing Project highlights a recent publication on “rendezvous methods" from Christopher Knight of Argonne National Laboratory and Steven Plimpton of Sandia National Laboratories. Check it out to learn how these algorithms work in the context of large-scale scientific modeling and simulation codes and how they can be help reduce performance bottlenecks and scale effectively to exascale machines.
ALCF's Corey Adams was one of six Argonne National Laboratory researchers to receive a DOE Early Career Research Program Award. Congrats to Corey and the 83 scientists across the country recognized with this prestigious award!
At the CogX Festival on June 16, 2021, ALCF's Venkat Vishwanath will join researchers from SambaNova Systems for a session titled "Supercharging science: How organisations can drive innovation with AI." The session will cover how ALCF is using novel AI architectures, like SambaNova, to scale AI model performance for science.
In addition to their roles at the ALCF, Michael Papka and Joe Insley serve as professors at Northern Illinois University. Students in their data visualization courses have been teaming up with WNIJ to help tell their stories with graphs, maps and other data visualizations.
This week's IEEE International Parallel & Distributed Processing Symposium (IPDPS) is a great place to hear about the latest research in parallel computing. If you're attending virtually, be sure to check out these paper presentations from ALCF researchers.
Researchers from Argonne National Laboratory and Illinois Institute of Technology were recognized with a best paper award at this week's IEEE/ACM CCGrid 2021 symposium for their study "DLIO: A Data-Centric Benchmark for Scientific Deep Learning Applications." The team developed a new benchmark for deep learning applications to emulate their data access patterns, identify I/O bottlenecks, and evaluate performance optimizations at scale with a diverse set of science applications on the ALCF's Theta supercomputer. Congratulations to the authors Hariharan Devarajan, Huihuo Zheng, Anthony Kougkas, Xian-He Sun, and Venkatram Vishwanath!
The 2021 Lustre User Group (LUG) conference will be held virtually May 19-20, 2021. The annual event will feature several talks on the latest Lustre developments, including "Lustre at Scale: ALCF Update” presented by ALCF's Alex Kulyavtsev and Gordon McPheeters on Wednesday, May 19.
In a recent study published in IEEE's Computing in Science and Engineering, Argonne National Laboratory's Murali Emani, Venkatram Vishwanath, Corey Adams, Michael Papka, and Rick Stevens collaborated with researchers from SambaNova Systems to explore how the ALCF's new SambaNova DataScale platform can help accelerate AI for science workloads.