The Theta system marks Cray and Intel’s first acceptance for a large-scale supercomputer featuring the latest generation of Intel® Xeon Phi™ processors formerly code named “Knights Landing.”
Theta has a peak performance of more than eight petaflops, and is currently running a broad range of scientific applications through the ALCF’s Theta Early Science Program. The Cray XC40 system was delivered to Argonne National Laboratory through a partnership with Intel as part of the DOE’s initiative to build state-of-the art supercomputers through the Collaboration of Oak Ridge, Argonne, and Lawrence Livermore National Laboratories (CORAL) program.
“The Argonne Leadership Computing Facility is committed to providing advanced supercomputing resources, like the Theta system, that help our researchers solve some of the world’s largest and most complex problems,” said Rick Stevens, Associate Laboratory Director for Argonne National Laboratory, in a statement. “We are pleased that Theta is now accelerating our scientific research.”
Cray, in partnership with Intel, will also deliver a flagship system called “Aurora” as part of the CORAL program. Aurora will be based on the next-generation Cray supercomputer code named “Shasta,” the successor to the current, industry-leading Cray XC™ line of supercomputers. Aurora is expected to have a peak performance of 180 petaflops and is scheduled for delivery in 2018.
“The Theta supercomputer at Argonne is an important step in the CORAL project and a milestone on the path towards the Aurora system,” said Dr. Rajeeb Hazra, Vice President of the Data Center group and General Manager of the Enterprise and Government group at Intel, in a statement. “The combination of Intel’s leading technology, latest Intel Xeon Phi processor technology, codenamed Knights Landing, Intel Scalable System Framework, and Cray’s system capabilities is helping to drive scientific discoveries and engineering breakthroughs.”