update performance and loss converging results #800
+71
−40
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
tianyu-l
added a commit
that referenced
this pull request
Jan 22, 2025
ghstack-source-id: c9eb138b7011e6ea99907c1aeee9ed0fda7d9b16 Pull Request resolved: #800
facebook-github-bot
added
the
CLA Signed
tianyu-l
commented
Jan 22, 2025
docs/performance.md
Outdated
|
|
||
| The experiments are conducted on NVIDIA H100 GPUs[^1] with 95 GiB memory, where each host is equipped with 8 GPUs and NVSwitch. Two hosts form a rack connected to a TOR switch. A backend RDMA network connects the TOR switches. |
There was a problem hiding this comment.
In the recent Meta CP paper (https://arxiv.org/abs/2411.01783), they mentioned:
We ran our performance benchmarks on the Grand Teton platform (Meta Engineering, 2022), where each host has 8 Nvidia H100 GPUs fully connected with NVLink (“host” and “node” are interchangeable in the subsequent text). Each H100 GPU is equipped with 96GB HBM2e with 2.4 TB/sec peak memory bandwidth. We tested on two subtypes of Grand Teton platforms: Grand Teton Training (GTT) and Grand Teton Inference (GTI). GTT hosts are inter-connected with backend RDMA network with 400 Gb/s per GPU, and GTI hosts are inter-connected with frontend network over TCP/IP with 100 Gb/s per GPU.
Would like to hear your thought on this @yifuwang
tianyu-l
added a commit
that referenced
this pull request
Jan 23, 2025
ghstack-source-id: ec5512945a3f156a989111b8226d274650eed4d2 Pull Request resolved: #800
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):