News from March 2016
Lawrence Livermore National Laboratory (LLNL) has purchased IBM Research’s supercomputing platform for deep-learning inference, based on 16 IBM TrueNorth neurosynaptic computer chips, to explore deep learning algorithms.
IBM says the scalable platform processing power is the equivalent of 16 million artificial “neurons” and 4 billion “synapses.” The brain-like neural-network design of the IBM Neuromorphic System can process complex cognitive tasks such as pattern recognition and integrated sensory processing far more efficiently than conventional chips, says IBM.
The technology represents a fundamental departure from computer design that has been prevalent for the past 70 years and could be incorporated in next-generation supercomputers able to perform at exascale speeds — 50 times faster than today’s most advanced petaflop (quadrillion floating point operations per second) systems.
- Published in