Ray Train DLC and EFA run PyTorch FSDP fine-tuning across two EKS nodes
Overview
AWS Machine Learning Blog describes a sample that fine-tunes Qwen2.5-1.5B with PyTorch Fully Sharded Data Parallel across eight NVIDIA L4 GPUs on two g6.12xlarge instances.
The job runs on Amazon EKS, with KubeRay managing a single RayCluster manifest that uses the published Ray Train Deep Learning Container without a custom image. Inter-node communication runs over Elastic Fabric Adapter (EFA).
Written by AI from the articles below · updated Oct 28, 3:00 PM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
- AWS Machine Learning BlogOfficialRay Train DLC and EFA run PyTorch FSDP fine-tuning across two EKS nodes
AIAWS Machine Learning Blog describes a sample that fine-tunes Qwen2.5-1.5B with PyTorch Fully Sharded Data Parallel across eight NVIDIA L4 GPUs on two g6.12xlarge instances. The job uses the Ray Train Deep Learning Container as published, with KubeRay running a single RayCluster manifest on Amazon EKS. Inter-node collectives run over Elastic Fabric Adapter (EFA), and no custom image is built.
Heat Trend
Not enough continuous observations to show a trend yet.