Armadillo is the newest addition to the shared campus HPC resource pool, providing CPU, GPU, large-memory compute nodes, and shared storage. Each GPU node has two Intel Xeon Gold 6530P processors and two NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, connected via Mellanox ConnectX-7 InfiniBand at 200 Gb/s. Each GPU node also provides 1.6 TB of local NVMe scratch storage.
Armadillo is jointly operated by HPE-DSI and the Research Computing  Data Core (RCDC), is housed in the RCDC data center within the AERB building, and entered production in August 2026. Allocations currently prioritize funded research projects; users must have a PI with an approved allocation. If you plan to use this system, please make sure your PI has been granted an allocation and you have requested an account on the system.  Please refer to the User Guide to find out about specifics of running jobs, selecting resources, and available software on this cluster.

The system comprises 95 compute nodes (63 CPU-only and 32 GPU nodes), delivering 6,080 CPU cores and 64 NVIDIA RTX PRO 6000 Blackwell GPUs with 6 TB of aggregate GPU memory. The system provides approximately 0.67 PetaFLOPS of theoretical FP64 and 8.0 PetaFLOPS of theoretical FP32 performance, with 1.54 million CUDA cores. Shared storage includes approximately 2.5 PB of HDD and 77 TB of flash storage.

Interconnect: Armadillo nodes are connected via Mellanox Infiniband NDR200 switches with a 200Gb/s Line Rate