In IBM simulates 530 billon neurons, 100 trillion synapses on supercomputer, Kurzweil AI reports that scientists announced the achievement of a milestone in the simulation of the human brain at a conference called SuperComputing 2012, last week:
Announced in 2008, DARPA’s SyNAPSE program calls for developing electronic neuromorphic (brain-simulation) machine technology that scales to biological levels, using a cognitive computing architecture with 1010 neurons (10 billion) and 1014 synapses (100 trillion, based on estimates of the number of synapses in the human brain) to develop electronic neuromorphic machine technology that scales to biological levels.”
IBM says it has now accomplished this milestone with its new “TrueNorth” system running on the world’s second-fastest operating supercomputer, the Lawrence Livermore National Lab (LBNL) Blue Gene/Q Sequoia, using 96 racks (1,572,864 processor cores, 1.5 PB memory, 98,304 MPI processes, and 6,291,456 threads).
IBM and LBNL achieved an unprecedented scale of 2.084 billion neurosynaptic cores* containing 53×1010 (530 billion) neurons and 1.37×1014 (100 trillion) synapses running only 1542 times slower than real time.
The scientists didn't say what their simulated brain was thinking about, but the picture above shows a map of the synaptic connections between simulated neurons in different regions of the simulated brain.
But, notice that even with the second fastest supercomputer in the world, the scientists were only able to simulate a sub-region of the human brain operating "only" 1542 times more slowly than your brain is operating at now as you read this, or when you are eating Hostess Twinkies while watching TV! This should make all couch potatoes feel superior.
Humans Rule! Still for a while longer, at least. But, that's the same thing the dinosaurs purportedly thought 260 million years ago.
End-of-post: Photos, video, and abstract footnotes below squiggle.
Here the authors exploring a group simulation of the human brain which my understanding is still legal in California.
And, Here is the Abstract from their Supercomputing 2012:
Inspired by the function, power, and volume of the organic brain, we are developing TrueNorth, a novel modular, non-von Neumann, ultra-low power, compact architecture. TrueNorth consists of a scalable network of neurosynaptic cores, with each core containing neurons, dendrites, synapses, and axons. To set sail for TrueNorth, we have developed Compass, a multithreaded, massively parallel functional simulator and a parallel compiler that maps a network of long-distance pathways in the Macaque monkey brain to TrueNorth. We demonstrate near-perfect weak scaling on a 16 rack IBM Blue Gene/Q (262144 CPUs, 256 TB memory), achieving an unprecedented scale of 256 million neurosynaptic cores containing 65 billion neurons and 16 trillion synapses running only 388× slower than real-time with an average spiking rate of 8.1 Hz. By using emerging PGAS communication primitives, we also demonstrate 2× better real-time performance over MPI primitives on a 4 rack Blue Gene/P (16384 CPUs, 16 TB memory).
And, here, is a painfully slow interview of Dr. Modha explaining their research.
6:47 PM PT: Hey, please check out my other posts of the last 24 hours: