November 14, 2018 – The first time Jibonananda Sanyal ever logged into a supercomputing system, the landing page told him he could use up to 500 cores, a modest number for some but a significant amount for the early career researcher.
“Up until then I had used two cores, at most,” he said. It wouldn’t be long before he multiplied that number many times over yet again, though.
“Fast forward a few years and I have used something to the effect of 130,000 processors on Titan to run some of our codes,” he said.
Rapidly scaling up seems to be a recurring theme for Sanyal, a researcher at the Department of Energy’s Oak Ridge National Laboratory. Since joining ORNL in 2011, Sanyal has been a postdoc, staff scientist and team lead for Scalable and High Performance Geocomputation. Today, he is acting leader of the Computational Urban Sciences group in ORNL’s Computational Sciences and Engineering Division.
“Being a group leader means a lot of hard work, but it is a challenge I thoroughly enjoy,” he said. “It is strategic and requires me to understand and define where we as a group will go. It is a wonderful opportunity and a little scary as well.”
At times, Sanyal wonders if going into science was a product of chance or an inevitability for him.
“When I was growing up back in India, at that time, you went into the sciences if you were reasonably decent in your studies because those were the sure-shot professions,” Sanyal said. “My dad was engineer who worked for the state electricity board and there were always technical discussions at home, so that environment steered me towards science.”
Sanyal was also interested in geography but the subject was clumped in with the humanities, so studying it at the same time as science wasn’t a viable option. Instead, he pursued degrees in computer science and engineering at St. Edmund’s College in Shillong and the University of Calcutta. He worked as a software engineer for a 3G mobile phone company for a time until his wife, a psychologist, wanted to pursue higher education and get her Ph.D. in the United States.
“That influenced me to the point where I basically decided it was time to do something more,” Sanyal said.
He began his doctoral work at Mississippi State University doing rapid prototyping for NASA Earth science systems, which opened a pathway back to the geographic and geospatial side of computing that had always interested him. He also worked with the National Oceanic and Atmosphere Administration and the National Weather Service’s Lower Mississippi River Forecast Center, where he first encountered high-performance computing.
Sanyal had several job opportunities available after graduation, but ORNL’s computing resources and his connection with a lab researcher through the IEEE Visualization conferences persuaded him to join up as a postdoc.
“The work at ORNL wasn’t directly aligned with what I had been working on during my Ph.D., but it was different and new enough to make me want to explore it,” he said. “Since then, it has been a journey.”
Making an impact. Sanyal is also the technical lead on EAGLE-I, an interactive energy infrastructure monitoring platform that allows agencies to track outages in near real-time and organize emergency responses. EAGLE-I was used by the Federal Emergency Management Agency and the U.S. Department of Homeland Security in the response and recovery efforts after hurricanes Harvey and Irma in 2017, where it was used to coordinate power restoration for hundreds of thousands of residents on the Gulf Coast.
“EAGLE-I was a great learning experience because it was at the heart of taking research into operations,” Sanyal said. “The mission of the program has been to support emergency response and being part of it has been a privilege.”
Recently, Sanyal has expanded his work with the Exascale Computing Project in urban systems science and is engaged in DOE’s SMART Mobility Laboratory consortium. The project uses advanced data analytics and data-derived simulations to help inform and control elements of an urban environment to alleviate energy bottlenecks like congestion. Sanyal hopes the project will be able to provide situational awareness of transportation in urban environments and help researchers use that information to make a real-world impact on the energy footprint of a city.
“There are very few projects where you get a direct connection to making a real difference, like with EAGLE-I,” Sanyal said. “For me to be able to provide to the needs of a community to accomplish a higher-level goal, that itself is a measure of how well we are doing.”
As time goes on and the projects grow larger and more complex, it becomes harder to see the “finish line” or find a satisfactory stopping point. The upside of this added complexity, though, is an increased awareness of the interdependencies between seemingly disparate factors and the ability to see how these different things amalgamate, Sanyal said.
“I want what I’m engaging with to succeed,” he said. "I don’t always know what the final outcome will be or what it may look like, but I try hard to understand what it could accomplish and what impact a particular engagement could make.”
Becoming a team and group leader has also changed how Sanyal sees his role as a researcher and collaborator. He only succeeds if his team succeeds, he said, and any leader that cares about their people will align the goals of each member with the long-term mission of the group and institution.
“I think it is a natural cadence of things, that as projects become bigger and more impactful, it’s not all about you,” Sanyal said. “It’s a group of people that accomplishes these things, and I’ve been privileged to work with smart people who have been stellar in the things they have done.”
Sanyal was lucky to have leaders and mentors who cared about the trajectory of his career and helped guide him through the big decisions. His advice to others looking for a mentor? Just ask. Early career researchers may be shy or fear being rejected, but it is worth trying, Sanyal said, as it is an invaluable opportunity to discuss challenges with someone who may have already navigated them in their own career. Today, Sanyal continues to do the same for up-and-coming researchers, offering mentorship at the lab and through his work with professional organizations.
“I engage with my community because I want to do it, not because it adds a line to my resume,” he said. “I want to contribute meaningfully to the scientific community by using the skills I have been educated in and grow the state of the science.”
UT-Battelle manages ORNL for the Department of Energy’s Office of Science. DOE’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States, and is working to address some of the most pressing challenges of our time. For more information, please visit science.energy.gov.