Leveraging Progress in Neurobiology for Computing Systems

Whether future technologies will be ultra-small CMOS transistors, nanotubes, or even individual molecules or bio-logical cells, these elementary components all share several common properties: they come in great numbers, they won’t be much faster or may even be way slower than current tran-sistors, they may be hard to precisely lay out and connect, and they may be faulty. The key question, then, is how can one design and, more importantly, program a computing sys-tem using billions of such components while we are not even capable of harnessing a few hundreds traditional cores? Now, when one considers these properties, it is almost ir-resistible to observe that nature has found a way to harness a huge number of elements with similar properties to real-ize complex information processing tasks. While suggesting that we design computing systems which somehow imitat