My Research Activities
My research interests lie broadly in designing and developing HPC-based parallel algorithms for biological systems and network science. For a complete list of my publications please visit my List of Publications.
HPC-Based Computational Biology
I work mainly on simulating enteric immune systems, where a wide variety of immune cells interact with each other to defend the body against foreign invaders like bacteria, microbes, viruses, toxins and parasites. Simulating such complex systems with millions to billions of cells requires not only apt mathematical formulation but also fast, high-performance computing systems.To address the issue, we designed and developed ENteric Immunity SImulator (ENISI), an agent-based simulator of the gastrointestinal immune mechanisms in response to invading pathogens. So far, I have published a book chapter, three journal papers, and a conference paper from this work. (Follow this link for more details.)
HPC-Based Algorithms for Network Science
I have designed, implemented, and analyzed MPI-based distributed memory parallel algorithms for generating massive random networks using various models that can generate networks with hundreds of billions of nodes and edges. I have designed and developed distributed memory-based parallel algorithms for the PA, CL, SB, BTER, and small-world models. To achieve the best performance, we have analyzed the nature of computational dependencies and developed novel, effective, parallel load balancing techniques. Therefore, our algorithms are highly scalable and can generate massive random networks. I have published another book chapter, a journal paper, and two conference papers in this work. Another article has been submitted recently to a top-tier conference. (Click this link for more details.)
I also work on Network Visualization. I designed and developed CINETViz, a visualization tool integrated with CINET, a HPC-based network analysis, and mining tool. I used Gephi, SigmaJS, and Apache Tomcat. It can visualizing graphs with thousands of nodes and edges in the web browser. I also used Google map based geographical maps. Here are few of my visualizations:
I also worked on other problems related to network mining and management. Mining and analyzing the network data requires high throughput, low latency data lookup. I designed and developed GRT (GPU-based Radix Tree), a new and highly efficient radix tree-based index-searching implementation for the GPU that supports both conventional point and range queries. GRT outperforms existing GPU-based index-searching system for a wide range of benchmarks. GRT achieves a high throughput of 106 and 130 million lookups per second for sparse and dense keys respectively, for point queries, with a large dataset of 64 million 32-bit keys. It yields 600 and 1000 million lookups per second for sparse and dense keys for range queries. A paper summarizing the work has been submitted to a top-tier conference. (Click this link for more details.)