Research synopsis: We use high-performance computing (HPC) to solve scientific problems that deal with sparse and unstructured data, like information about river flow, energy distribution, and biological networks. Our group conduct research to make graph analysis scalable and efficient. Our research covers all layers of graph analytics, graph machine learning, sparse matrix algorithms, parallel algorithms, high-performance software and libraries and applications from biology, earth science and scientific computing. In developing these algorithms, we aim to push the boundaries of problem scalability to the limits of the world’s most powerful computing systems. Throughout the development of these algorithms, we also strive to create general-purpose solutions. This ensures their applicability across various scientific domains, allowing them to effectively tackle a wide range of scientific problems, even when the data looks very different.

We have opening for multiple postdoctoral positions. Please contact us to know more. 

We always welcome motivated PhD applicants with research interests in high-performance computing, graph analysis, sparse-matrix computations, graph machine learning, and related problems. 

Research Sponsors

 

National Science Foundation - Wikipedia        SC Logos | U.S. DOE Office of Science (SC)        ECP     Georgia Tech Awarded IARPA Contract to Evaluate Emu Technology System | David A. Bader           CICP Foundation, Inc. - Idealist