Virginia Tech®home


ARC offers a variety of compute resources for Virginia Tech researchers and departments, including CPU, GPU, and high-throughput computing clusters. Learn more about each of our compute clusters below:

TinkerCliffs came online in the summer of 2020, and it is our largest CPU cluster with nearly 42,000 compute cores and over 93 TB of RAM.  It is housed in the Steger Hall HPC Data Center on the main Blacksburg campus.

Infer came online in January of 2021 and provides 18 nodes, each with an Nvidia T4 GPU. This cluster is a great all-purpose resource for researchers who are making their first forays into GPU-enabled computations of any type.

Cascades is a 236-node system capable of tackling the full spectrum of computational workloads, from problems requiring hundreds of compute cores to data-intensive problems requiring large amount of memory and storage resources.

DragonsTooth is a 48-node system designed to support general batch HPC jobs. Released to the Virginia Tech research community in August 2016, DragonsTooth was reprovisioned with Slurm as its scheduler as a replacement for Moab/Torque in late 2018.

Huckleberry is a high performance computing system targeted at deep learning applications. It consists of two login nodes and Fourteen IBM Minksy S822LC compute nodes.  Note: Huckleberry is scheduled to be retired in spring of 2022; users should consider another of ARC's GPU resource for deep learning applications.