diff --git a/micro-benchmarks/nccl-tests/README.md b/micro-benchmarks/nccl-tests/README.md index bbd5ee9e..8f9fead4 100644 --- a/micro-benchmarks/nccl-tests/README.md +++ b/micro-benchmarks/nccl-tests/README.md @@ -37,7 +37,7 @@ The NCCL tests are packaged in a container. > |-----------------------|-------------|---------------------------------------------------------------------------------------------| > |`GDRCOPY_VERSION` | `v2.4.1` | [link](https://github.com/NVIDIA/gdrcopy) | > |`EFA_INSTALLER_VERSION`| `1.35.0` | [link](https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/efa-start.html#efa-start-enable) | -> |`AWS_OFI_NCCL_VERSION` | `v1.12.0-aws`| [link](https://github.com/aws/aws-ofi-nccl) | +> |`AWS_OFI_NCCL_VERSION` | `v1.12.1-aws`| [link](https://github.com/aws/aws-ofi-nccl) | > |`NCCL_VERSION` | `v2.23.4-1` | [link](https://github.com/NVIDIA/nccl) | > |`NCCL_TESTS_VERSION` | `v2.13.10` | [link](https://github.com/NVIDIA/nccl-tests) | @@ -45,7 +45,7 @@ The NCCL tests are packaged in a container. 1. Build the container image with the command below: ```bash EFA_INSTALLER_VERSION=1.35.0 - AWS_OFI_NCCL_VERSION=v1.12.0-aws + AWS_OFI_NCCL_VERSION=v1.12.1-aws NCCL_VERSION=v2.23.4-1 NCCL_TESTS_VERSION=v2.13.10 docker build -f nccl-tests.Dockerfile \ @@ -82,7 +82,7 @@ To run the NCCL tests on EKS, you will need to build the container image, then p 1. Create the ECR repository if it does not exist ```bash EFA_INSTALLER_VERSION=1.35.0 - AWS_OFI_NCCL_VERSION=v1.12.0-aws + AWS_OFI_NCCL_VERSION=v1.12.1-aws NCCL_VERSION=v2.23.4-1 NCCL_TESTS_VERSION=v2.13.10 ECR_REPOSITORY_NAME="nccl-tests" diff --git a/micro-benchmarks/nccl-tests/buildspec.yaml b/micro-benchmarks/nccl-tests/buildspec.yaml index 8a064f1b..18ea9350 100644 --- a/micro-benchmarks/nccl-tests/buildspec.yaml +++ b/micro-benchmarks/nccl-tests/buildspec.yaml @@ -4,7 +4,7 @@ env: variables: GDRCOPY_VERSION: "v2.4.1" EFA_INSTALLER_VERSION: "1.35.0" - AWS_OFI_NCCL_VERSION: "v1.12.0-aws" + AWS_OFI_NCCL_VERSION: "v1.12.1-aws" NCCL_VERSION: "v2.23.4-1" NCCL_TESTS_VERSION: "v2.13.10" exported-variables: diff --git a/micro-benchmarks/nccl-tests/nccl-tests.Dockerfile b/micro-benchmarks/nccl-tests/nccl-tests.Dockerfile index 4aa71c03..e3e32e58 100644 --- a/micro-benchmarks/nccl-tests/nccl-tests.Dockerfile +++ b/micro-benchmarks/nccl-tests/nccl-tests.Dockerfile @@ -4,7 +4,7 @@ FROM nvidia/cuda:12.2.2-devel-ubuntu22.04 ARG GDRCOPY_VERSION=v2.4.1 ARG EFA_INSTALLER_VERSION=1.35.0 -ARG AWS_OFI_NCCL_VERSION=v1.12.0-aws +ARG AWS_OFI_NCCL_VERSION=v1.12.1-aws ARG NCCL_VERSION=v2.23.4-1 ARG NCCL_TESTS_VERSION=v2.13.10