I am a graduate student with the Department of Computer Science and Engineering at The Ohio State University. I am working towards my PhD in high-performance computing. My co-advisors are Dr. P. Sadayappan and Dr. Mario Lauria.
Biographical
Information
I grew up in Jamshedpur, a small city in eastern India. I received my undergraduate degree from the Birla Institute of Technology and Science, Pilani and my Masters from the University of Cincinnati. My interests include among other things, music (Jazz, Rock, Heavy Metal), fishing and swimming. I also enjoy Latin-American dancing (Salsa, Merengue & Bachata).
I grew up in Jamshedpur, a small city in eastern India. I received my undergraduate degree from the Birla Institute of Technology and Science, Pilani and my Masters from the University of Cincinnati. My interests include among other things, music (Jazz, Rock, Heavy Metal), fishing and swimming. I also enjoy Latin-American dancing (Salsa, Merengue & Bachata).
Research
My research interests span the broad areas of computer architecture, operating systems and high-performance computing. In the past, I have worked on the memory bandwidth issues related to CMPs. My current research focuses on two related areas: parallel file systems and parallel I/O over wide-area networks.
Parallel File Systems: As storage systems evolve, the block-based design of today's disks is becoming inadequate. As an alternative, Object-Based Storage Devices (OSDs) offer a view where the disk manages data layout and keeps track of various attributes about data objects. By moving functionality that is traditionally the responsibility of the host OS to the disk, it is possible to improve overall performance and simplify management of a storage system. The capabilities of OSDs will also permit performance improvements in parallel file systems, such as further decoupling metadata operations and thus reducing metadata server bottlenecks. My work examines the feasibility of OSDs for use in parallel file systems, in particular, discovering techniques to accommodate this high-performance usage model. I am currently investigating multiple aspects of the mismatch between the needs of a parallel file system, specifically PVFS2 and the capabilities of OSDs. Some of the open research topics in this area include mapping data to objects, metadata management, transport, caching and reliability. This work is being done in collaboration with researchers at the Ohio Supercomputer Center.
Parallel I/O over Wide-Area Networks: One of the challenges in high-performance computing is to provide users with reliable, remote data access in a distributed, heterogeneous environment. The increasing popularity of high-speed wide-area networks and centralized data repositories lead to the possibility of direct, high-speed access to remote data sets from within a parallel application. My research involves investigating mechanisms to increase the available I/O bandwidth between remote clusters over wide-area networks. I work on parallel I/O libraries (MPI-IO) and the Storage Resource Broker file system to find efficient ways to perform remote I/O in a high-performance computing environment.