Purdue's Anvil cluster built in partnership with Dell and AMD consists of 1,000 nodes with two 64-core AMD EPYC "Milan" processors each and delivers over 1 billion CPU core hours each year, with a peak performance of 5.1 petaflops. Each of these nodes has 256GB of DDR4-3200 memory. A separate set of 32 large memory nodes has 1TB of DDR4-3200 memory each. Anvil's nodes are interconnected with 100 Gbps Mellanox HDR100 InfiniBand.

16 nodes each with four NVIDIA A100 Tensor Core GPUs providing 1.5 PF of single-precision performance to support machine learning and artificial intelligence applications.

The Purdue Anvil AI system has 21 nodes each with four NVIDIA 80GB H100 SXM GPUs to support machine learning and artificial intelligence applications.

The Anvil Notebook Service is a highly scalable, AI-enabled platform for web-based Jupyter notebooks. The service has access to 3rd generation AMD EPYC CPUs and NVIDIA H100 GPUs with support for dynamic allocation of portions of vRAM and GPU cores to notebook instances, making the service ideal for providing many concurrent interactive AI programming environments. Each notebook session is integrated with Jupyternaut, a generative AI tool that provides chat-based coding assistance to users from open-source models via AnvilGPT, or from commercial providers like OpenAI, Gemini and Anthropic. Note: Users are required to bring their own subscriptions and API keys for commercial providers. The Anvil Notebook Service is also fully integrated with the Anvil NAIRR HPC resource, providing the ability to submit Slurm jobs to Anvil’s batch system from within notebooks for additional access to GPUs and has access to various storage tiers such as home, project, and scratch. Users can create their own notebook images with custom software and use them on the platform as well as run custom software via containers (apptainer) inside a notebook session.