One of the most powerful supercomputers in the world, Japan’s K computer, has for the first time successfully modeled human brain activity. Clocking in at over 10 petaflops, the K computer is the world’s fourth fastest. Yet even with 705,024 processor cores and 1.4 million GB of RAM, a single second’s worth of neural activity took 40 minutes to calculate.
Researchers from the RIKEN HPCI Program for Computational Life Sciences, the Okinawa Institute of Technology Graduate University (OIST) in Japan and Forschungszentrum Jülich in Germany all worked together to carry out the largest general neuronal network simulation ever performed.
Crucial to the experiment was open-source simulation software, called Neural Simulation Technology or NEST, which harnessed 82,944 processors of the K system to reproduce a complex network of 1.73 billion nerve cells connected by 10.4 trillion synapses. Despite the size of the studied network, it represents only 1 percent of the brain’s entire neural network.
The study was undertaken to test the limits of simulation technology and the capabilities of the K machine. Now that researchers have a better idea of what’s possible, they can use this experience to guide future brain simulation studies.
The advanced simulation also provides a glimpse as to what the future holds. Exascale computing will enable unprecedented science. Whole organ modeling of the heart, brain, etc. — these will become a reality. The hope is that such research will lead to treatments and cures for debilitating diseases like Parkinson’s and Alzheimer’s.
As project researcher Markus Diesmann affirms: “If peta-scale computers like the K computer are capable of representing 1% of the network of a human brain today, then we know that simulating the whole brain at the level of the individual nerve cell and its synapses will be possible with exa-scale computers hopefully available within the next decade.”