This 'brain-inspired' supercomputer will explore deep learning for the U.S. nuclear program

Based on IBM's TrueNorth chip, it will have the equivalent of 16 million neurons and 4 billion synapses but consume energy like a tablet PC

Comments

A new low-power, "brain-inspired" supercomputing platform based on IBM chip technology will soon start exploring deep learning for the U.S. nuclear program.

Lawrence Livermore National Laboratory announced on Tuesday that it has purchased the platform, based on the TrueNorth neurosynaptic chip IBM introduced in 2014. It will use the technology to evaluate machine-learning and deep-learning applications for the National Nuclear Security Administration.

The computer will process data with the equivalent of 16 million neurons and 4 billion synapses and consume roughly as much energy as a tablet PC. Also included will be an accompanying ecosystem consisting of a simulator; a programming language; an integrated programming environment; a library of algorithms and applications; firmware; tools for composing neural networks for deep learning; a teaching curriculum; and cloud enablement.

A single TrueNorth processor consists of 5.4 billion transistors wired together to create an array of one million digital neurons that communicate with one another via 256 million electrical synapses.

With 16 TrueNorth chips, the new system will consume a mere 2.5 watts of power, allowing it to infer complex cognitive tasks such as pattern recognition and integrated sensory processing far more efficiently than conventional chips can, IBM said.

TrueNorth was originally developed under the auspices of the Defense Advanced Research Projects Agency’s (DARPA) Systems of Neuromorphic Adaptive Plastic Scalable Electronics (SyNAPSE) program in collaboration with Cornell University.

Lawrence Livermore will collaborate on the technology with IBM Research, universities and other partners within the Department of Energy.

Neuromorphic computing will have a role in Lawrence Livermore's national security missions and could change how the lab does science, according to Jim Brase, deputy associate director for data science with Lawrence Livermore.