University of Cambridge > > Computer Laboratory Computer Architecture Group Meeting > Communication Challenges for Extreme-Scale, Real-Time Neural Network Simulation

Communication Challenges for Extreme-Scale, Real-Time Neural Network Simulation

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Prof Simon Moore.

Neural network simulation is an example of an embarrassingly-parallel problem with complex communication patterns. The neurons being simulated are inherently parallel but the spike messages between them exhibit complex patterns. We consider GPUs as a possible simulation platform, but discover that their methods of processing and communication make them unsuited to the task. As an alternative we propose a simulation platform consisting of a network of Field Programmable Gate Arrays (FPGAs), connected using integrated high-speed serial transceivers. This system is much more suited to the complex communication patterns and real-time requirements of an extreme-scale neural network simulation.

This talk is part of the Computer Laboratory Computer Architecture Group Meeting series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2022, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity