4 ms·
CUDA already has this built in (for many years) using UVM. That won't take away the bottleneck of the transfer, but it does make it transparent to the user by h
by shaklee3 7y ago
CUDA already has this built in (for many years) using UVM. That won't take away the bottleneck of the transfer, but it does make it transparent to the user by handling all the paging/copies.
- mzs 7y agoI thought that just made GPU memory map equal to CPUs' and you need much for VM (like paging).