Argonne Announces Call for Papers for the Exascale Aurora System

Winning projects will get started early with preparing to work at exascale on Aurora, which will be finished in 2021.

The Argonne Leadership Computing Facility (ALCF), which is part of the US Department of Energy, is accepting proposals for data science and machine learning projects for ALCF’s Aurora supercomputer, which is scheduled to debut in 2021 as the world’s first exascale computer system — capable of performing a quintillion calculations per second. According to the press release, Aurora will have more than 50,000 nodes and more that 5 Petabytes of total memory.

The Aurora Early Science Program (ESP) is intended to let scientists get started early with preparing applications, simulations, and research projects to run on the Aurora system when it is ready. ALCF will award 10 projects in the areas of data and learning. According to the announcement, proposals should have “... strong aspects of data science (Big Data, data-intensive computing, experimental/observational/simulation data analytics, etc.) and/or machine learning (deep learning, neural networks, discovery of patterns and reduced models for scientific data and/or simulation modeling, etc.).”

Project proposals are due by April 8, but that’s “before midnight in any time zone anywhere on Earth,” so you’d better shoot for April 7 to be on the safe side. See the announcement at the ALCF website for more information. You can also create an account at Easychair for Aurora ESP Data and Learning 2018 to learn more about Aurora ESP and the project proposal process.