(www.tomshardware.com)

521 points hd4 | 1 comments | 20 Oct 25 12:31 UTC | HN request time: 0.193s | source

1. ibejoeb ◴[20 Oct 25 17:51 UTC] No.45646911[source]▶

Sounds like this virtual GPU is a separate scheduler. I wonder what kind of latency is introduced by marshaling all that data around.

Alibaba Cloud says it cut Nvidia AI GPU use by 82% with new pooling system