Training program

Participants from around the world learn supercomputing skills during the annual Argonne training program

In recent years, for example, ATPESC has included an entire day dedicated to training participants in the use of machine learning and deep learning methods for science.

“Too often, students and post-docs join research groups and accumulate ad hoc knowledge of high performance computing that primarily focuses on methods already in use by their group,” Loy added. “ATPESC offers exposure to a very wide range of topics and serves as a core program. It plays an important role in priming the pipeline of the next generation of computer scientists. Some of our participants went on to lead important IT projects at DOE or became professors who in turn sent their students to CLTSP. We even had CLTS alumni as speakers this year.

One of those alumni was Suyash Tandon, a software systems design engineer at AMD. He attended CLTSP last year and returned this year as a speaker.

“ATPESC is a great place for the scientific computing community to meet and learn from each other,” Tandon said. “The ATPESC workshop provides attendees with the ‘bleeding advice’ of developments in the HPC field, and they can ‘dip their hands in the fat’ with hands-on sessions.”

As part of the hardware architectures training, participants learned about DOE’s future exascale supercomputers as well as cutting-edge AI platforms, including Cerebras, SambaNova, Groq and Habana systems deployed in Argonne. ATPESC has enabled DOE computing facilities and AI companies to connect and better inform the community about emerging technologies that are redefining scientific computing.

“We hope attendees learned the differentiation and value that SambaNova’s comprehensive software and hardware solutions bring to performing deep learning and large-scale AI for scientific applications with ease. use at the highest levels of performance, precision and scale, ”said Marshall. Choy, vice president of products at SambaNova.

While this year’s virtual program was safe for participants during the pandemic, Loy and his team hope to be able to return in person next year when CLTSP marks its 10th anniversary.

“While we have continually adjusted and updated the program, we will be doing a more in-depth review over the coming year,” Loy said. “In addition, we are considering ways to involve the growing number of alumni of CLTSP, which now stands at over 700 in total. “

ALCF, OLCF, and NERSC are user facilities of the DOE Office of Science located in Argonne, Oak Ridge, and Lawrence Berkeley National Laboratories, respectively.

View original with photos here:


The Argonne Leadership Computing Center provides high-performance computing capabilities to the scientific and engineering community to advance fundamental discovery and understanding across a wide range of disciplines. Supported by the Office of Science of the United States Department of Energy (DOE), the Advanced Scientific Computing Research (ASCR) program, the ALCF is one of two dedicated DOE leadership computing facilities in the country. to open science.

Argonne National Laboratory seeks solutions to urgent national problems in science and technology. The country’s leading national laboratory, Argonne conducts cutting-edge fundamental and applied scientific research in virtually all scientific disciplines. Argonne researchers work closely with researchers from hundreds of companies, universities, and federal, state, and municipal agencies to help them solve their specific problems, advance U.S. scientific leadership, and prepare the nation for a better future. With employees from over 60 countries, Argonne is managed by UChicago Argonne, LLC for the Office of Science of the US Department of Energy.

The Office of Science of the United States Department of Energy is the largest proponent of basic physical science research in the United States and strives to address some of the most pressing challenges of our time. For more information, visit https: // ener gy .gov / s c ience.

Source: Argonne Leadership Computing Facility

Source link

Leave a Reply

Your email address will not be published. Required fields are marked *