MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUDA Moat Still Alive

Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by ryao on Monday December 23, 2024 @12:41PM (#42496133)

Infiniband is very much an industry standard:

https://www.infinibandta.org/member-listing/

As far as I know, everyone is free to sign up with the infiniband trade association and implement the specification.

Furthermore, if you use RDMA over Ethernet, you are using infiniband at a low level. RoCE which enables it was originally called Ethernet over infiniband. It is maintained by the infiniband trade association.

Omnipath was Intel’s failed effort to try to kill an open standard. It purchased QLogic’s infiniband business, killed it in favor of omnipath and sold it when it failed.

[ Parent | on HN ]

Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by mrlongroots on Monday December 23, 2024 @12:50PM (#42496173)

> Furthermore, if you use RDMA over Ethernet, you are using infiniband at a low level. RoCE which enables it was originally called Ethernet over infiniband.

This is wrong.

RDMA over Ethernet is... RDMA over Ethernet. There is no Infiniband involved.

RoCE was motivated by supporting RDMA, which was then an IB-only feature, over regular Ethernet. The user-level APIs are the same (verbs), but the underlying architecture is all different --- it is traditional Ethernet with link-level flow control to make it lossless (pause frames).

[ Parent | on HN ]

Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by ryao on Monday December 23, 2024 @01:18PM (#42496417)

This is not what I have been told by others.

[ Parent | on HN ]

Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by mrlongroots on Tuesday December 24, 2024 @12:59AM (#42499997)

Tip: copy-paste our entire exchange as a Gemini prompt and ask "who's correct?".

(I did that yesterday and was very satisfied :) )

[ Parent | on HN ]

Re:MI300X vs. H100 vs. H200 Benchmark Part 1: Training – CUD... by ryao on Tuesday December 24, 2024 @02:29PM (#42504270)

You are not the first person to say “the AI assistant liked what I said over what you said” to me on the Internet, to which I say they will tell you whatever you want to hear.

I am not interested in continuing this discussion, but if you want to do your own research, I suggest starting with the fact that the Infiniband Trade Association controls the RoCE specification. I suggest you avoid using LLMs, for obvious reasons.

[ Parent | on HN ]