Upcoming 16-Core AMD CPU Already Chosen by Supercomputers

AMD made an official announcement in which it reveals that both existing and upcoming processors are being used in high-performance computing applications.

Since HPC applications, supercomputers as it were, have become the core of any and all research efforts, AMD, naturally, wants to get contracts for as many as possible.

Its latest press release says that this goal was achieved in at least some measure.

HECToR is one of the supercomputers based on AMD chips, Operated by the University of Edinburgh, UK, and funded by the Engineering and Physical Sciences Research Council.

Another is CSCS, Located in Manno, Switzerland, as is NAMEM (Ulaanbaatar, Mongolia) and HLRS, the High Performance Computing Center at the University of Stuttgart (Germany).

These, along with conglomerates from Cray Inc. (Titan) are making liberal use of AMD Fusion APUs (accelerated processing units) and AMD Opteron 6100 Series CPUs (central processing units).

What's more, AMD already reached agreements involving the upcoming "Interlagos" chip with 16 cores.

"ORNL is deploying more than 25,000 of AMD’s ‘Interlagos’ processors over the next few months as we upgrade Jaguar to the new Titan system," said Buddy Bland, project director of ORNL's Leadership Computing Facility.

"Our users are excited about the increase in performance over previous-generation processors, and our sponsors are delighted with the power savings that will make Titan one of the world’s most powerful and efficient research tools."

For people who want to know of more recent developments, Penguin Computing made the first AMD A8-3850 APU-based cluster in Sandia National Labs, Albuquerque, NM.

"HPC is not a one-size-fits-all environment, and requires new technologies to keep pace with customer demands," said Paul Struhsaker, corporate vice president and general manager, Commercial Business at AMD.

"Whether it’s our upcoming 'Interlagos' processor or our energy-efficient APU, AMD’s unique x86 and world-class graphics IP place us at the heart of some of the fastest systems as we push well beyond the petaflop towards the exaflop."