Wyatt Madej presented on the college’s access to a wider range of computing resources.
ANGELINA PERSAUD/THE QUADRANGLE
By Angelina Persaud, Managing Editor/Asst. News Editor
Manhattan College’s Information Technology Services (ITS) recently held an event hosted by Wyatt Madej to expand on the campus’ computing infrastructure, including the ScienceDMZ connected network.
The hybrid event was hosted by Wyatt Madej, the technical architect in the ITS department, where he explained the key ideas and processes that allow students to have access to a wider range of computational resources. Previously, Madej was also one of the primary authors for the college’s largest National Science Foundation (NSF) grant, which awarded MC a two-year grant valued at $639,778 for advancing campus computing power.
He began with a welcome and introduction to some of the conceptual thoughts that are the driving force behind the newly expanded campus infrastructure. Cyberinfrastructure (CI) refers to the computing systems, software, storage systems and other high-performance compute resources used for educational purposes at the college.
Madej explained that the core of advanced cyberinfrastructure is to allow students to have more hands-on experience learning about research from a more connected network.
“Using CI, we might be able to reduce the time it takes to make discoveries, which is really important,” Madej said. “More time discovering, less time doing the computation. Just to increase that discovery means more time learning, and it exposes students to CI and the research community; exposing them to things that have resources they might find useful after their time with us.”
Madej also briefly noted that the previous NSF grant to promote the ScienceDMZ at the college is beginning to take effect with new hardware being swapped out of various labs in Leo and the Research and Learning Center (RLC).
According to manhattan.edu, “The project also funds the design, implementation and purchase of specially built data transfer nodes connected to the Science DMZ that will be used to help move massive amounts of data between Manhattan College researchers and their collaborators. The project will provide College researchers and educators a connection to local, state and national research networks such as NYSERNet and Internet2, providing the community unprecedented access to scientific research networks.”
The “nodes” refer to a specific component of computer hardware where data can be processed and distributed. Prior to the grant, MC was not able to access a large magnitude of compute power as these nodes are typically occupied by larger institutions on the network.
The NSF website described the ScienceDMZ as a way to facilitate further research opportunities for both students and faculty while allowing them access to other universities’ computational networks for a more interconnected experience.
“This Science DMZ utilizes recognized best practices for research cybersecurity, balancing network performance and security, and establishes data transfer nodes to facilitate data transfers between Manhattan College researchers and their collaborators,” the grant states.
The grant also includes the goal to provide higher quality educational opportunities for students that they can use beyond their undergraduate years at the college.
“This network will support and grow these activities to advance research and STEM education programs that educate students, contribute to new scientific discoveries, inform public policy, facilitate institutional partnerships, and provide a better STEM education to train the next generation of scientists and engineers,” the grant stated.
As a result of the ScienceDMZ coming to fruition, the college joined the Open Science Grid (OSG), a larger university network for research, where students are able to produce more results for their specific research interests as well as collaborate on projects with other researchers.
“The Open Science Grid, it’s really beneficial for those that use free open-source software,” Madej said. “And most importantly, it’s free. There is no charge to U.S.-based academic researchers to use this resource. It’s best suited for doing image analysis, such as analyzing MRI images, geographic information system images, and mapping stuff.”
Madej noted that there is an application process to gain access to the OSG where students or faculty can submit an application with their specific research requests and be connected to an entry point to start using the grid.
He also noted that the college now has access to more machine learning and supercomputing resources, courtesy of the National Research Platform, which includes access to compute power from graphic processing units (GPUs).
“Some really talented folks over there [at the National Research Platform] found out ways to do amazing research on GPU resources,” Madej said. “The workloads here are mostly focused towards computer science and AI research, but it is not limited to those. We have a great team of facilitators who can help get those workloads up and running and help you get your computation happening on their platform. Currently, as we stand, there’s a total of 1,500 single-precision GPUs and 10 petabytes of storage.”
As Madej noted, CI is increasingly evolving over the years and will continue to impact student’s educational and research goals.
“We’re seeing, across the world, a huge surge in the understanding of cyberinfrastructure, whether it be arts, social science, humanities, those folks are becoming users of CI, too,” Madej said.
Wanda Parisien is a computing expert who navigates the vast landscape of hardware and software. With a focus on computer technology, software development, and industry trends, Wanda delivers informative content, tutorials, and analyses to keep readers updated on the latest in the world of computing.
