Skip to content

Use better shape count for NVIDIA megatron training. #1855

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Jul 2, 2025
Merged

Use better shape count for NVIDIA megatron training. #1855

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 9 additions & 2 deletions ...rastructure/ai-infra-gpu/ai-infrastructure/nemo-megatron-training-oke/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@ on the Oracle Container Engine for Kubernetes (OKE) using
Reference results from NVIDIA to train Llama 3 can be found on the
[NGC Catalog](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/dgxc-benchmarking/resources/llama3-dgxc-benchmarking).

Reviewed: 18.03.2025
Reviewed: 01.07.2025

# When to use this asset?

Expand All @@ -31,7 +31,14 @@ This guide is loosely based on the
[to the instructions](https://github.com/oracle-quickstart/oci-hpc-oke/tree/main#instructions-for-deploying-an-oke-cluster-with-gpus-and-rdma-connectivity),
importing one of the images and creating a GPU partition with BM.GPU.H100.8 nodes.

The configuration here assumes a minimum of 16 BM.GPU.H100.8 nodes.
The configuration here assumes a minimum of 1 BM.GPU.H100.8 node for
training with 8B parameters, and a minimum of 8 BM.GPU.H100.8 nodes for 70B
parameters.

If another shape is used, the NCCL and MPI parameters in the Kubernetes
[configuration map](./files/training/templates/mpi.yaml) should be adapted
using the same parameter values as the
[performance testing scripts](https://github.com/oracle-quickstart/oci-hpc-oke/tree/main/manifests/nccl-tests).

- Ensure that the follwing setting is selected under the "OKE Cluster" section:

Expand Down