Anvil

Purdue University will soon be the home of Anvil, a powerful new supercomputer that will provide advanced computing capabilities to support a wide range of computational and data-intensive research spanning from traditional high-performance computing to modern artificial intelligence applications.

News & Events

  • No articles or events found at this time.

Overview

Anvil, which is funded by a $10 million award from the National Science Foundation (NSF), will significantly increase the capacity available to the NSF's Extreme Science and Engineering Discovery Environment (XSEDE), which serves tens of thousands of researchers across the U.S., and in which Purdue has been a partner for the past nine years. Anvil will enter production in 2021 and will serve researchers for five years. Additional funding from the NSF will support Anvil's operations and user support.

The name "Anvil" reflects the Purdue Boilermakers' strength and workmanlike focus on producing results, and the Anvil supercomputer will enable important discoveries across many different areas of science and engineering. Anvil also will serve as an experiential learning laboratory for students to gain real-world experience using computing for their science, and for student interns to work with the Anvil team for construction and operation. We will be training the research computing practitioners of the future. Learn more about Anvil's mission in the Anvil press release.

Anvil is funded under NSF award number 2005632. Carol Song is the principal investigator and project director. Preston Smith, executive director of Research Computing, Xiao Zhu, computational scientist and senior research scientist, and Rajesh Kalyanam, data scientist, software engineer, and research scientist, are all co-PIs on the project.

Scientific Highlights

CPU and GPU nodes

Anvil’s CPU and GPU nodes will help accelerate structural predictions and simulations to uncover protein structures of the novel SARS-COV-2 or antibody binding sites in viruses (Richard Kuhn, Purdue Institute of Inflammation Immunology and Infectious Disease)

Large Memory Nodes

Anvil’s large memory nodes and interactive computing capability will help develop and execute big data applications for research and training in Translational Omics (training materials from NIH Big Data to Knowledge, BD2K program, Min Zhang, Purdue Center for Cancer Research)

Complex Physics Models

Anvil’s large number of high-core count CPU nodes will enable and accelerate the kind of mid-scale fluid computations necessary to develop and run complex physics models, such as turbulence models for simulating aircraft-vortex collisions (Carlo Scalo, Xinran Zhao, Purdue Mechanical Engineering, Aeronautics and Astronautics)

Data Science Tools

Anvil’s composable infrastructure and data science tools will enable the deployment of an AI-based Recommender Engine for Intelligent Transient Tracking that can coordinate and strategize follow-up tracking based on near real-time processing of observatory data (Dan Milisavljevic, Purdue Physics and Astronomy)

Interactive Applications

Anvil’s CPU cluster and interactive applications will aid in the multifaceted simulation and visualization of casting processes from the iron and steelmaking industries; which can help address furnace control, stability, and material and energy efficiency considerations (Chenn Zhou, Purdue Center for Innovation through Visualization and Simulation)

Open OnDemand

Anvil’s Open OnDemand interactive applications and large memory nodes will enable the use of photogrammetry and drone mapping Windows desktop tools to construct 3D models of archaeological sites for anthropological research and education (Pix4DMapper generated 3D model using imagery from a Bronze Age fortress in Armenia, Ian Lindsay, Purdue Anthropology)

Specifications

Anvil will be built in partnership with Dell and AMD. All Anvil nodes will have 128 processor cores, 256 GB to 1 TB of RAM, and 100 Gbps Infiniband interconnects.

Anvil Front-Ends
Front-Ends Number of Nodes Processors per Node Cores per Node Memory per Node Retires in
4 Two 3rd Gen AMD EPYC™ CPUs @ 2.0GHz 128 256 GB 2026
Anvil Sub-Clusters
Sub-Cluster Number of Nodes Processors per Node Cores per Node Memory per Node Retires in
A 1,000 Two 3rd Gen AMD EPYC™ CPUs @ 2.0GHz 128 256 GB 2026
B 32 Two 3rd Gen AMD EPYC™ CPUs @ 2.0GHz 128 1 TB 2026
C 16 Two 3rd Gen AMD EPYC™ CPUs @ 2.0GHz + Four NVIDIA A100 GPUs 128 256 GB 2026

Anvil nodes will run CentOS 8 and use Slurm (Simple Linux Utility for Resource Management) as the batch scheduler for resource and job management. The application of operating system patches will occur as security needs dictate. All nodes will allow for unlimited stack usage, as well as unlimited core dump size (though disk space and server quotas may still be a limiting factor).

Documentation

Coming Soon

We're working hard to write up an in-depth guide that will walk you through getting access, submitting jobs, and dealing with common tasks such as installing needed software or libraries.

Contact Us

For inquiries concerning Anvil, email us at anvil@purdue.edu.

For alternate means of communication or inquiries concerning other Research Computing resources and services, please refer to our Contact page.


Anvil is supported by the National Science Foundation under Grant No. 2005632.