The Pittsburgh Supercomputing Center has won a $5 million award from the National Science Foundation to build Neocortex, an AI supercomputer that incorporates the Cerebras Systems Wafer Scale Engine technology introduced last year along with Hewlett Packard Enterprise’s shared memory Superdome Flex hardware.
PSC, a joint research organization of Carnegie Mellon University and the University of Pittsburgh, said the Cerebras–HPE system will be available by the end of this year.
The Neocortex architecture will incorporate two Cerebras CS-1 AI servers, each powered by a Cerebras Wafer Scale Engine (WSE) processor designed for faster deep learning training and inferencing that Cerebras said is the world’s largest computer chip. The WSE contains 400,000 AI-optimized cores implemented on a 46,225 square millimeter wafer with 1.2 trillion transistors. When announced last August, Cerebras said the WSE is 56.7 times larger than the largest GPU, which measures 815 square millimeters and 21.1 billion transistors. The WSE also contains 3,000 times more high speed, on-chip memory, and has 10,000 times more memory bandwidth, according to Cerebras.
Cerebras Wafer Scale Engine
The Cerebras servers will be coupled with a HPE Superdome Flex provisioned with 24 terabytes of memory, 205 terabytes of high-performance flash storage, 32 powerful Intel Xeon CPUs, and 24 network interface cards for 1.2 terabits per second of data bandwidth to each Cerebras CS-1, according to PSC. Neocortex will use the HPE hardware to “enable flexible pre- and post-processing of data flowing in and out of the attached Cerebras WSEs, preventing bottlenecks and taking full advantage of the WSE capability.”
Neocortex, planned to be available free of charge to researchers before the end of this year, will support the most widely used deep learning frameworks and automatically accelerate them, for improved ease of use, PSC said. The machine will be federated with PSC’s new Bridges-2 supercomputer, also to be installed in 2020, “to form a singularly powerful and flexible ecosystem for high performance AI, data analytics, modeling and simulation,” PSC said.
The Neocortex project is led by Paola Buitrago, PI and PSC director of AI and Big Data.
Today’s news comes on the heels of several other recently announced NSF supercomputing awards, including those at NCSA, Purdue University and Indiana University.
Here’s more information on Neocortex and the NSF award.
