aws-efa
CommunityEnable fast multi-node GPU training with EFA.
Authortylertitsworth
Version1.0.0
Installs0
System Documentation
What problem does it solve?
AWS Elastic Fabric Adapter (EFA) provides a high-bandwidth, low-latency network path for distributed GPU workloads by enabling SRD/libfabric-based OS-bypass and GPUDirect RDMA, reducing CPU overhead and enhancing cross-node communication on EKS.
Core Features & Use Cases
- EFA-enabled multi-node NCCL communication for GPU clusters on Kubernetes.
- NCCL integration and performance tuning for GPUDirect RDMA workloads.
- Troubleshooting guidance for topology, security groups, and placement groups in AWS GPU deployments.
Quick Start
Create an EFA-enabled EKS nodegroup and verify NCCL throughput using the aws-ofi-nccl provider.
Dependency Matrix
Required Modules
None requiredComponents
references
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: aws-efa Download link: https://github.com/tylertitsworth/skills/archive/main.zip#aws-efa Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.