HPC focuses on developing software and tools that improve the performance, scalability and programmability of parallel computing systems and applications. Recent years have seen a sharp rise in the demand for computational resources and there are many diverse areas, including simulation and modeling in engineering and design, data mining, social network computations, bioinformatics, network security and graph computations that require computational power far greater than what a single processor can provide. While there have been rapid advances in parallel computer hardware, much research and development is needed on the software front to exploit these new hardware models. The aim of the HPC lab is to bridge the gap between the parallel hardware and software.
Current research directions include load balancing of irregular applications, fine-grain parallel algorithms for large-scale applications and scalability of message passing middleware.