The National Science Foundation has recently awarded three research grants for the 2007 proposal competition Foundations of Computing Processes and Artifacts (CPA) to CSE faculty D. K. Panda, Srinivasan Parthasarathy and Xiaodong Zhang. The three research projects will address three different fundamental problems in computer systems and architecture.
Professor DK Panda is leading a collaborative project entitled "Designing Next Generation Communication and I/O Subsystems with Multi-core Architecture" with Pavan Balaji at the University of Chicago. His research will investigate issues in designing the following components for next generation HEC systems: Multicore-aware Message Passing Interface (MPI), enhanced MPI with dedicated communication threads, multicore-aware I/O subsystem and reliability and fault tolerance.
The project led by Srinivasan Parthasarathy is entitled "Scalable Data Analysis: An Architecture Conscious Approach." He seeks to employ an architecture-conscious approach to scalable data analysis on modern cluster systems interconnected through a high speed network. The central thesis of his work is that current day algorithms for data analysis often grossly under-utilize resources provided by such systems. The project seeks to address this limitation in the context of key application drivers drawn from scientific simulations, bioinformatics and homeland security.
Xiaodong Zhang is leading another NSF CPA project entitled "Algorithm Design and Systems Implementation to Improve Buffer Management for Fast I/O Data Accesses." He and his collaborator, Professor Song Jiang at Wayne State University, will address the increasingly more serious problem of "disk wall" by efficiently improving and enhancing the memory caching management in operating systems.