Tag Archives: featured

Welcome to Advanced Research Computing (ARC)

Providing high performance computing systems and services for Virginia Tech students and faculty

What is ARC?

Advanced Research Computing (ARC) provides centralized research computing infrastructure and support for the Virginia Tech research community. ARC's resources include high-performance computing systems, large-scale data storage, visualization facilities, software, and consulting services, and are available to faculty and students across all disciplines. We welcome researchers of all experience levels.

How can ARC help me achieve my research goals?

ARC's high-performance computing (HPC) systems can help researchers process big data sets, run deep learning applications, conduct visualization, store data sets, and perform other tasks that are too large or complex to do on a local computer.

For example, researchers often come to ARC if they have:

  • Very large jobs to run (e.g., high resolution or large-scale datasets)
  • Many jobs to run (parameter sweeps, many datasets)
  • Need for specialty hardware (GPU, memory, high bandwidth storage, fast network, scale)

Additionally, ARC's resources for data storage and sharing can make it much easier for researchers from different disciplines or in different locations to collaborate on a project.

How to get started with ARC

If you are interested in using ARC's resources for your current or future projects, or if you would just like to learn more about our computing systems and services, please request a consultation or drop by our office hours.

You do not need to have any prior experience with high-performance computing — our team can assist you in determining the right system for your project. We also offer introductory training sessions throughout the year via the Professional Development Network, and our computational scientists are available for classroom presentations on HPC.

Is there a cost to use ARC's systems?

ARC's systems and services are available to all Virginia Tech faculty at no cost. However, researchers and groups have the option to add compute costs to grants or contracts through ARC's Cost Center, which allows usage of additional compute or storage resources for a fee. Departments and faculty can also purchase priority access to an ARC system for up to five years through ARC's Investment Computing Program.

Helpful resources for new ARC users:

Infer, New AI/ML GPU Cluster

ARC is proud to announce the release of a new cluster called Infer, which provides 18 Intel Skylake nodes each equipped with an Nvidia T4 GPU. The cluster’s name "Infer" alludes to the AI/ML inference capabilities of the T4 GPUs derived from the "tensor cores" on these devices. We think they will also be a great all-purpose resource for researchers who are making their first forays into GPU-enabled computations of any type. For more information about the T4 architecture on T4 architecture and performance relative to, e.g., V100 GPUs, see the following pages:

Cluster details and examples are provided on our Infer page. Users may login at infer1.arc.vt.edu and try it out at their earliest convenience. For now, software installs are mostly limited to CUDA and associated toolchains; if users have additional requests please, they may submit them via Help ticket.

New ARC flagship CPU cluster Tinkercliffs featured in VT news

Tinkercliffs, ARC's most recent cluster, is online and is making news owing to its scale (more than 300 nodes and almost 42,000 CPU cores) and some of the challenges the acquisition and commisioning faced in the COVID-19 era.

See the news article here:
https://vtnews.vt.edu/articles/2020/10/it-arc-tinkercliffs-supercomputer.html

ARC also has a webpage for introducing Tinkercliffs to the VT community and will continue to add more resources in the coming months:

https://arc.vt.edu/tinkercliffs/

As of early October, access to the cluster is provided though a consultation with an ARC computational scientist. You may reach out to one of the team directly or submit a ticket requesting a consultation via arc.vt.edu/support.