AMD Accelerator Cloud
The AMD Accelerator Cloud (AAC) is a private hosted platform that offers users access to the latest AMD hardware resources, tools, and ready-to-use application software. This platform facilitates the rapid and cost-effective development of solutions using AMD GPUs.
On this website, you will find comprehensive documentation on how to use the AMD Accelerator Cloud.
📚 Plexus platform
Plexus is a web-based platform which simplifies access to and usage of the infrastructure available in AAC. It provides a ready-to-use set of applications for the wide range of AMD accelerators, including solutions for AI/ML workloads, high-performance computing, data analytics, and visualization.
🆕 What's new in Plexus
Stay up to date with the latest features, improvements, and bug fixes!
Latest Release: Version 1.16.0
🚀 Getting started
New to AMD Accelerator Cloud? Start here:
- What's Plexus - Learn about the Plexus platform
- Create an Account - Get started with your AAC account
- What's a Plexus Workload - Understand how workloads work
- Launch One-click Workload - Run your first workload quickly
Explore Plexus Documentation →
🖥️ Bare Metal access
For advanced users with direct SSH access to AMD clusters. This section provides guides on connecting to clusters, using container tools, and running HPC/AI applications directly.
Popular Topics:
Explore Bare Metal Documentation →
Note: This is for users who have been granted direct SSH access and shared their public keys with the AAC team. Most Plexus users access workloads through the web platform or on-the-fly SSH connections.
- Bare metal prerequisites – Access, partition naming, ROCm, and Podman
📖 Glossary
Definitions of common terms (AAC, Plexus, Slurm, partition, queue, ROCm, HIP, workload, and related commands). Link to the glossary from other docs instead of repeating definitions.
❓ Frequently asked questions (FAQs)
Quick answers to common questions about using the AMD Accelerator Cloud platform.
Common Questions:
- How to launch/run a workload?
- Why is my workload in pending state?
- How to SSH into a container?
- How to connect JupyterLab?
🔗 Quick links
- What's New - Latest Release
- Run a Workload
- Monitor a Workload
- PyTorch Application Guide
- TensorFlow Application Guide
💬 Need help?
- Documentation Questions: Browse the sections above or use the search bar
- Technical Support: Contact your AMD sponsor
- Report Issues: Reach out to your AMD sponsor with details