From <Bytes> to Breakthroughs

Texas A&M University-San Antonio is developing an innovative new resource designed to enhance cutting-edge research initiatives and foster departmental collaboration.

The High-Performance Computing Research Center (HPCRC) will equip researchers and students with the right tools to carry out scientific studies utilizing massive data sets and substantial computational power.

The center will also enable users to connect and collaborate with other researchers on a broad range of projects. As part of this effort, students have the opportunity to learn marketable skills like statistical analysis and coding, while earning degrees in the natural and life sciences, among many other fields. 

Dr. Izzat Alsmadi, a professor in the Department of Computational, Engineering and Mathematical Sciences, is the project’s principal investigator. He and members of his team are organizing a series of HPCRC workshops and training sessions.

Last spring, Alsmadi and Dr. Donna Lehman, a professor of human genomics in the Department of Health and Behavioral Sciences, co-hosted a cross-disciplinary workshop that featured faculty and researchers from various colleges discussing research computing needs and use cases.

“The HPCRC creates opportunities for interdisciplinary collaboration in fields such as business, cybersecurity, biology and education,” said Alsmadi. “This reflects a deliberate push toward building collaborative, cross-college research projects that leverage shared high-performance computing infrastructure to support diverse scholarly work.”

“We’ve needed this resource for a long time,” Lehman added. “I am looking forward to connecting with other faculty through this resource to initiate new approaches to analyzing data.”

Drs. Yulun Han and Utpal Smart, both assistant professors in the Department of Natural Sciences, attended the spring workshop and said they’re eager to utilize the full resources that the HPCRC will offer.

“As an evolutionary biologist, I work with very large sets of salamander genomics data,” said Smart. “An HPC will enable me to better analyze this information.”

Han’s research focuses on computational simulations of light–matter interactions. His projects can generate terabytes of data and require advanced quantum chemistry methods that demand substantial computing power.

"We've needed this resource for a long time. I am looking forward to connecting with other faculty through this resource to initiate new approaches to analyzing data." ~ Dr. Donna Lehman

“To carry out my research, I rely heavily on HPC resources, including computer nodes capable of handling large-scale, parallel simulations, as well as storage systems that can support terabyte-scale data,” he said. “The continued investment in and development of our HPC capabilities will be essential for my work.”

Dr. Burak Aksoylu, an associate professor of mathematics in the Department of Computational, Engineering and Mathematical Sciences, was a panelist at the spring workshop. He said he is planning to use HPC resources for his research on domain decomposition methods, which involves dividing a complex problem's domain into smaller, more manageable sub-domains.

“This will bring a lot of savings in computational time and resources,” he said.    

Alsmadi stresses the project is still a work in progress and fully developing this new computer infrastructure will take time and campus-wide coordination.     

“There’s also a substantial learning curve to utilize all the different resources,” he said. “It's not like using a typical laptop.”   

The HPCRC is supported through a $500,000 startup fund from Texas A&M University System Chancellor's Research Initiative, as well as a $100,000 grant from the National Science Foundation’s Campus Cyberinfrastructure program, which invests in campus-level cyberinfrastructure improvements and innovations.

With training now underway, Alsmadi said the project will be developed in multiple phases, as he and his team will next work with other faculty members to develop strategies for implementation and campus-wide use.

The University’s HPCRC—essentially a collection of sophisticated servers—is currently located in the Central Academic Building. A small number of faculty and students are using the center, with more expected as training progresses. Alsmadi said there are also plans to develop a more expansive center in one of the new campus buildings scheduled to open in the coming years.

Alsmadi explains that rather than a standard central processing unit, the HPCRC uses powerful computers that have a graphics processing unit, which is designed to handle complex mathematical calculations along with graphics, video and images.

“A process that would take days to complete with a normal laptop can be done in hours or even minutes with an HPC,” he said.   

Lehman, who is the project’s co-principal investigator, explains that as a professor of human genomics, she has many ongoing studies that generate and analyze extremely large datasets.

“We have 30 years' worth of clinical data from thousands of individuals that we have to keep stored in a safe, compliant way,” she said. “We have to be able to access that data and run intensive programs to analyze it. We're going to be relying on the HPCRC to do that kind of work. I think it's going to be pretty astounding.”

She noted that the key to moving the project forward is to engage faculty and other potential users, including students, to get their perspectives on what they need.

Lehman indicated that data science is used in nearly every field nowadays, and the HPCRC innovations can be used to pique students’ interest in research and develop peer-teaching opportunities.  

“Field students from genomics, biology or chemistry will work with computer science students and teach each others,” she said. “And these students can then introduce faculty to each other, who can start their own collaborative projects. It’s a bottom-up approach and a different level of learning.” 

Like Lehman, Dr. Davida Smyth, a professor of biology in the Department of Natural Sciences, is also excited about the HPCRC’s potential. She said she needs access to enhanced storage and capacity for her research and classes, including virology and bacteriology courses in which the students generate and analyze microbiome data.

“We're building the capacity to do this kind of work at a centralized campus location,” she said. “It’s going to be really amazing.”   

Smyth said as the HPCRC training continues, it’s critical to identify potential users’ specific needs.  

“With an initiative like this, we need to identify the platforms and capabilities that people are looking for,” she said. “I think it's a remarkable opportunity for us to show a common need across campus and across disciplines. By working together, we can be more strategic and less wasteful in terms of time and resources.”

Moreover, it’s an opportunity to better serve students as the University builds on existing programs in AI, cybersecurity and machine learning. 

“Now we can integrate high-performance computing into biology, chemistry and other fields,” she said. “As we consider the entire educational pipeline, from K-12 up to when our students are ready to graduate, the HPCRC will help build critical skills over time and better prepare our students to enter the workforce.”