Anvil has a tough name for a tough job. Named for the Purdue Boilermakers, Anvil’s moniker promises results and, true to its name, it delivers. Funded by the National Science Foundation (NSF) and now officially part of ACCESS, Anvil has enabled important science and engineering discoveries since 2020.
Two current research projects – a study on plants and another on cancer – have used Anvil to run intensive simulations to get results. This range of computational and data-intensive research demonstrates the importance of supercomputing resources such as Anvil.
Photosynthesis and Sustainability
A research team, led by Josh Vermass from Michigan State, was allocated time on Anvil to study photosynthesis and plant metabolism. The goal of the research was to gain insights that could help the researchers facilitate more efficient energy conversion. If a plant efficiently converts solar energy into plant growth, it can help keep food production more sustainable for Earth’s growing population. If plants become more efficient at growing and producing, it could mean improvements in plant growth – from shorter growth cycles to more robust produce.
For their study, the researchers used Anvil to run a number of molecular-scale simulations. These atomic simulations enable them to see the process of a plant absorbing solar energy and converting it for use in growth. By running various simulations, they can learn exactly how the molecules, proteins and other nanostructures within the plant work to convert the energy. This introduces the possibility for researchers to identify areas where they might introduce modifications to a plant at a cellular level to improve the photosynthesis process, making the plant grow faster with the same or even fewer resources.
Our new molecular simulation engines can make GPUs talk to each other, and how fast a simulation can go is limited by how much they can communicate. The key fact of Anvil for us is that it has four A100 GPUs. They’re faster than the GPUs we have locally, and because they’re all linked together, we can use all of them at once.
Josh Vermass, Professor of Biochemistry and Molecular Biology, Michigan State University
You can read more about this work on Anvil here (published September 9, 2022): Anvil Accelerates Sustainability and Photosynthesis Research
Training Cancer Researchers
From studying energy conversion in plants to training students, professors and practicing physicians in cancer research, Anvil serves as a resource.
Min Zhang, a professor of statistics at Purdue, illustrates how supercomputers can be used for more than just crunching numbers. Since 2020, the National Cancer Institute has funded a series of workshops for training in cancer research. This year Zhang was allocated space on Anvil, which enabled him and his team to invite more participants than organizers originally thought could be accommodated. “There’s no way we could do this for so many people without Anvil,” said Zhang.
Due to how research time was allocated in previous workshops, researchers were restricted to one node of a cluster provided by Purdue’s Rosen Center for Advanced Computing. With Anvil, a new allocation process was chosen by workshop organizers, and a total number of hours was allotted to the project, which allowed for more simultaneous use.
Students also reported that Anvil was “very fast and easy to use.”
“We gave the students the freedom to do real practice with real big data.”
Doug Crabill, senior academic IT specialist for the department of statistics, Purdue University
You can read more about this story here (published September 8, 2020): Anvil Used to Train Cancer Researchers in Big Data Analysis
These are just two examples of how ACCESS-allocated resources can help accelerate research. If you have a research project you think might benefit from time on a supercomputer or via any of the cyberinfrastructure resources available through the ACCESS program, we encourage you to explore our Knowledge Base or reach out to the Allocations team and they’ll help you find the right resource. Remember, the open allocation period for large projects runs through Oct. 15, 2022.
Project Details
Institution: ITaP Research Computing
University: Purdue University
Funding Agency: National Science Foundation
Grant Number: 2005632
This material is based upon work supported by the National Science Foundation under Grant 2005632. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.