Undergraduate Topics:
Graduate Topics :
OpenMP(Open Multi Processing):- OpenMP is an API that supports multi-platform shared memory multiprocessing programming in C, C++, and Fortran. OpenMP uses a portable, scalable model that gives programmers a simple and flexible interface for developing parallel applications for platforms ranging from the standard desktop computer to the supercomputer.
CUDA(Compute Unified Device Architecture):- NVIDIA CUDA (Compute Unified Device Architecture) technology is the world's only C language environment that enables programmers and developers to write software to solve complex computational problems in a fraction of the time by tapping into the many-core parallel processing power of GPUs.
Software Piracy:- Software piracy as “the illegal copying, distribution, or use of software.” Piracy includes casual copying of particular software by an individual orbusiness. With the advancement of technological tools software piracy has increased worldwide. This presentation and Report includes causes and effects of software piracy.
Multi threaded Asynchronous Graph Traversal for In-Memory and Semi-External Memory:- The paper discusses three graph traversal algorithms and their asynchronous parallel approach.
(a) Breadth First search
(b) Single Source Shortest Paths
(c) Connected Components
Direction-Optimizing breadth First Search :- To optimize breadth first search on large graphs, we implemented a hybrid approach which uses conventional top-down approach with the novel bottom-up approach. Initially during the top-down phase, nodes in the active frontier(queue) searches for a non-visited child while during the new bottom-up approach, the non-visited children look for a parent in active frontier. This hybrid approach reduces the number of edges to be examined and thus provides the required speedup.
Concurrent Read/Write with RCU and RLU :- The paper discusses the concurrent readers with simultaneous writer problems. In read copy update(RCU), we demonstrated that how multiple readers can work in parallel with a concurrent writer on the same data object. Similarly, in read log update(RLU), we discussed that how multiple readers can work with multiple writers on the same data object.
Energy Characterization in Handheld Devices :- Paper discusses the mobile GPU's and how the GPU's can be used optimally in handheld devices where power source is limited wireless battery. Some specific uses of mobile GPU's are discussed like face recognition and how the GPU cores can be fully utilized to give same performance as desktop GPU's with same accuracy and minimal power consumption. Paper also discusses the parallel frame rendering approach in which GPU cores are divided in two disjoint sets of cores and we make them to process consecutive frames in parallel.