Hacker News Reader

Parent Full thread

BiasRegularizer

Also, GPU has VERY limited memory and no virtual memory.

View on HN

> VERY limited memory and no virtual memory

This is incorrect.

Compared to 2-4TB max in a 2 socket server system, the 16 or 32GB/card you get on a GPU is pretty small.

The other issue is loading that memory - PCIe 4 is still the transfer time bottleneck between GPU and main memory.

The part about no virtual memory is kind of incorrect though. It doesn't work exactly like on the CPU, but still.

I'm not totally up to speed on modern GPU's - do they have a MMU that handles page faults on non-resident pages like a CPU would?

My understanding is that the GPU memory models are different enough that what an OS traditionally calls "virtual memory" couldn't be implemented in the same way.

Reply on news.ycombinator.com