Interim Chair and Professor Srinivas Aluru and CSE Professor and Associate Chair Ümit V. Çatalyürek have both been inducted into the 2020 Class of Society of Industrial and Applied Mathematics (SIAM) Fellows.
Nominated for exemplary research and outstanding service to the SIAM community, Aluru and Çatalyürek’s nominations account for two of the 28 inducted into this year’s international fellows program.
Ümit V. Çatalyürek
Çatalyürek is being recognized by SIAM for his contributions to the fields of combinatorial scientific computing, and high-performance and parallel algorithms – fields of research in which he has won a number of awards in prior.
This award listing includes a National Science Foundation (NSF) CAREER award and a Class of 2016 IEEE Fellow for his research contributions to the fields of discrete algorithms and high-performance computing.
“Discrete algorithms, and in particular, graph and hypergraph algorithms, are my passion. When solving some of the problems, I feel like a child trying to solve a puzzle,” he said.
“To solve larger problems, we need parallel algorithms because the problems either do not fit on a single computer or we need to make the solutions more efficient and useful in real-life settings.”
This idea of solving for these larger problems is where combinatorial scientific computing comes into play. This interdisciplinary research area uses graphs and parallel algorithms to solve computational science and engineering problems on large-scale HPC architectures. Combinatorial problems arise in many different fields of science, such as when constructing the genome of a new species that is being sequenced for the first time.
“At the core of almost all large-scale scientific computing is the question of how to partition and assign work and computation to processors of a large complex parallel system, and how to orchestrate the execution to minimize execution time, or energy consumption. These are some of the fundamental questions I try to answer in my research,” said Çatalyürek.
One recent project of Çatalyürek’s that addresses these questions this is a three-year, cross-institute NSF project that aims to lower the barrier to entry for software engineers developing new high-performance applications on large scale parallel systems.
Aluru is being recognized by SIAM for his contributions to the field of computational genomics with sequential and parallel discrete algorithms research and for his leadership in data science and engineering.
He is the recipient of numerous awards for his research, including being selected as an IEEE Fellow in recognition for high-performance computing (HPC) research and being selected as an American Association for the Advancement of Science Fellow in recognition of research at the intersection of computing and biology.
“These are all various aspects of the same interdisciplinary research my group has been focusing on for 23 years now,” he said.
According to Aluru, “Awards such as SIAM Fellow are given for sustained contributions over a very long period of time; sometimes, you have to wait to see the impact of a work. One major example of this impact within my group is the work we did to develop parallel algorithms and software for assembling genomes from tens to hundreds of millions of genomic fragments.”
The specific research project Aluru references enabled the successful sequencing of the maize genome, the first plant genome ever to be sequenced. The sequence was published in a 2009 Science paper that is now, according to Google Scholar, cited 3271 times, and is a go-to resource for biological research on this important food crop.
“Maize is much more complex than a human genome, with about 75 percent of the genome consisting of repeats of various sizes, tripping up efforts in accurately assembling the genome,” he said.
Before the success of Aluru’s team, it was believed that sequencing genomes on highly distributed memory-challenged platforms, such as the IBM Blue Gene/L supercomputer, was not feasible. However, this was not Aluru’s only time overcoming the impossible in this field of research. He led the first group to develop efficient parallel discrete algorithms which allowed for biological networks to be constructed at the whole genome scale within minutes using the world's fastest supercomputers.
“It is really a tribute to the hard work of many of my graduate students who contributed greatly to my work over the years. Hopefully, the elevated recognition for our group's work will also enhance their opportunities,” said Aluru.
In addition to his leadership and research within CSE, Aluru maintains a number of roles to enhance data science initiatives at the state and national levels. Currently, he serves as the Executive Director of the Institute for Data Engineering and Science (IDEaS), co-leads the NSF South Big Data Regional Innovation Hub which nurtures big data partnerships between organizations in the 16 Southern States and Washington D.C., as well as the NSF Transdisciplinary Research Institute for Advancing Data Science.