Real time Spaun on SpiNNaker - Functional brain simulation on a massively-parallel computer architecture
Andrew Mundy · Research Portal (King's College London) · 2017
Model building is a fundamental scientific tool. Increasingly there is interest in building neurally-implemented models of cognitive processes with intention of modelling brains. However, simulation of such models can be prohibitively expensive in both time and energy required. For example, Spaun - the world's first functional brain model, comprising 2.5 million neurons - required 2.5 hours of computation for every second of simulation on a large compute cluster.SpiNNaker is a massively parallel, low power architecture specifically designed for simulation of large neural models in biological real time. Ideally, SpiNNaker could be used to facilitate rapid simulation of models such as Spaun. However Neural Engineering Framework (NEF), with which Spaun is built, maps poorly to architecture - to extent that models such as Spaun would consume vast portions of SpiNNaker machines and still not run as fast as biology. This thesis investigates whether real time simulation of Spaun on SpiNNaker is at all possible.Three techniques which facilitate such a simulation are presented. The first reduces memory, compute and network loads consumed by NEF. Consequently, it is demonstrated that only a twentieth of cores are required to simulate a core component of Spaun network than would otherwise have been needed. The second technique uses a small number of additional cores to significantly reduce network traffic required to simulated this core component. As a result simulation in real time is shown to be feasible. The final technique is a novel logic minimisation algorithm which reduces size of routing tables which are used to direct information around SpiNNaker machine. This last technique is necessary to allow routing of models of scale and complexity of Spaun. Together these provide ability to simulate Spaun model in biological real time - representing a speed-up of 9000 times over previously reported results - with room for much larger models on full-scale SpiNNaker machines.