Colocating ML Inference and Training with Fast GPU Memory Handover.

Jiali Wang, Yankui Wang, Mingcong Han, Rong Chen

Browse the full USENIX paper archive.