In this video, University of Colorado Boulder doctoral student Eric Wolf describes how the Janus supercomputer runs climate simulations that help him study why the early earth was warm despite the Sun being much less luminous that it is today. This quandary is known as the Faint Young Sun Paradox. As one of the first container-based supercomputers in academia, the Janus supercomputer comprises 1368 compute nodes, each containing 12 cores, for a total 16,416 available cores.
“The team at the Graduate School of Computer and Information Sciences wanted to give the computer a nickname – a name that would not only convey the super computer’s enormity in terms of size; but a name that would definitively link the super computer to NSU. Our mascot is the “shark” and the Megalodon, is the largest prehistoric shark known to man.”
“Computers are becoming an increasingly cheaper, more powerful tool that cannot be ignored by professionals. Computer simulation reproduces the behavior of natural and man-made systems to help us understand, predict, and communicate. In this series kick-off, we will show you how computer simulation is used by LLNL scientists on the worlds fastest computers. We will also show you how you can get started doing your own computer simulations with free, open-source tools for class projects or just for fun.”
“The main topics for our April 7-9 meeting in Santa Fe are industrial partnerships with large HPC centers and how they’re working, with perspectives from the U.S., France and the UK. We’ll also take another hard look at what’s happening with processors, coprocessors and accelerators and at potential disruptive technologies, as well as zeroing in on the HPC storage market and trends and the CORAL procurement that involves Oak Ridge, Argonne and Livermore.”
“Systems like Argonne’s Mira, an IBM Blue Gene/Q system with nearly a million cores, can enable breakthroughs in science, but to use them productively requires expertise in computer architectures, parallel programming, mathematical software, data management and analysis, performance analysis tools, software engineering, and so on. Our training program exposes the participants to all those topics and provides hands-on exercises for experimenting with most of them.”
Fred Streitz from Lawrence Livermore National Lab presented this talk at the Stanford HPC Conference. “The HPC Innovation Center bridges American industry’s growing need for advanced solution to complex challenges in research and development with LLNL’s forefront supercomputing and disruptive scientific capabilities.”