Overflowing emerging neural network inference tasks from the GPU to the CPU on heterogeneous servers.

Adithya Kumar, Anand Sivasubramaniam, Timothy Zhu

Browse the full SYSTOR paper archive.