←back to thread

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

(www.tomshardware.com)
521 points hd4 | 1 comments | 20 Oct 25 12:31 UTC | HN request time: 0s | source
Show context
muddi900[dead post] ◴[20 Oct 25 14:43 UTC] No.45644484[source]▶
[flagged]
Yoric ◴[20 Oct 25 16:07 UTC] No.45645504[source]▶
I haven't read the paper yet, but it's here: https://ennanzhai.github.io/pub/sosp25-aegaeon.pdf aka https://dl.acm.org/doi/10.1145/3731569.3764815 .

So, definitely not state media, probably not lying on the fundamentals. Of course, still presumably viewed favorably by the CCP, I'd imagine.

replies(1): >>45646380 #
muddi900 ◴[20 Oct 25 17:11 UTC] No.45646380[source]▶
Well if it is real, we will surely see out Claude Code limits go back up.
replies(1): >>45648372 #
1. Yoric ◴[20 Oct 25 19:47 UTC] No.45648372[source]▶
From the abstract, this seems to be a scheduling mechanism for datacenters that serve multiple models. I have no idea whether this applies to Anthropic.