Supercomputer Art Connects Research to Australian Roots at Pawsey

For our customers, owning a Cray supercomputer brings a lot of pride. It’s easy to understand why. They’re making incredible discoveries with them. In a lot of ways, a supercomputer becomes part of an organization’s “family.” We see it time and time again in the thought and attention our customers put into designing art for their system cabinet fronts. Pawsey Supercomputing Centre is one of these organizations. A national high-performance computing facility located in Perth, Western Australia, the Pawsey Centre supports almost 200 research projects and 1,500 researchers in solving problems as varied as ending famine to understanding the nature of the universe. Supplying the HPC power are two Cray® XC™ series supercomputers — ... [ Read More ]

Cray Announces Third Exascale Supercomputer Win

Clean sweep, three-peat, triple crown… call it what you will. At Cray we call it an immensely proud day. On August 13 we announced our third contract in six months to deliver an exascale supercomputer to the U.S. Department of Energy. That’s right. Three. Named El Capitan and slated for delivery to Lawrence Livermore National Laboratory, this third system ($600 million total contract value) follows two previously announced Cray and DOE exascale supercomputers — Aurora at Argonne National Laboratory and Frontier at Oak Ridge National Laboratory. The trio of sister systems will be the first exascale supercomputers in the United States. And all three will be built on the new Cray® Shasta™ architecture. When it comes online in ... [ Read More ]

DOE and Cray Announce Exascale Supercomputer Frontier

Here at Cray, we’re charging into the exascale era. We followed March’s announcement of the United States' first exascale system ― Argonne National Laboratory’s Aurora supercomputer ― with another last week. At an Oak Ridge National Laboratory-hosted event, the U.S. Department of Energy (DOE) revealed Frontier, an even more powerful exascale supercomputer expected to land at ORNL in 2021. Frontier will be based on our new Shasta™ architecture and Slingshot™ interconnect and will feature high-performance AMD EPYC™ CPU and AMD Radeon Instinct™ GPU technology. And the stats on the system are impressive: 1.5 exaflops of performance (more operations per second than 160 of today’s top supercomputers ... [ Read More ]

How Hyperparameter Optimization Improves Machine Learning Accuracy

In January, a team of Cray developers and researchers published a paper, “Recombination of Artificial Neural Networks,” on arXiv.org, highlighting the hyperparameter optimization (HPO) capability Cray announced in November. We cover their findings in this blog post. Using a variety of high-performance computing systems and neural network models, the Cray team demonstrated that the hyperparameter optimization capabilities introduced in the Cray® Urika®-CS and Urika®-XC AI and analytics software suites improve the time-to-accuracy as well as final accuracy of machine learning models trained on Cray systems. The table below, excerpted from the paper, highlights the improvements achieved using Cray’s HPO capability across a range of ... [ Read More ]

Meet Slingshot: An Innovative Interconnect for the Next Generation of Supercomputers

Today Cray is announcing our next-generation “Shasta™” supercomputer, featuring our latest generation of scalable interconnect, code-named “Slingshot.” As a long-time network architect, I’m pretty excited. Slingshot is our 8th major generation of scalable HPC network, and there have been some great milestones along the way. We started back in 1992 with the Cray T3D, Cray's first massively parallel system. Implemented in BiCMOS, its network latency was just 12 ns per hop. That was followed in 1996 by the pioneering Cray T3E system, which had the first-ever implementation of adaptive routing in an HPC network (by a long shot!). In 2005, Cray pioneered the design of high-radix switches. Our YARC switch for the Cray X2 implemented ... [ Read More ]