About the CUDA Research Center at Wright State University


  At Wright State University several faculty members are exploring approaches to accelerate scientific computing, data intensive computing, and computer security and cryptanalysis using GPU computing. Specifically, our research team is developing a compiler framework based on polyhedral model to optimize the input GPU kernel programs, to automatically insert synchronization statements into GPU kernels at compile time, while minimizing the number of inserted statements. Our research team has developed a new thread scheduling technique that simultaneously exploits inter-warp and intra-warp locality of the GPU kernel programs and also automatically coalesces memory. 

  In June 2013, Wright State University was selected as a CUDA Teaching Center.  As part of this program, Dr. Meilin Liu developed an undergraduate/graduate level course “Parallel Programming on many-core GPUs” (CS 4370/6370), and Dr. Jack Jean has integrated CUDA and GPU computing into his courses. As a CUDA Teaching Center, the Department of Computer Science and Engineering at Wright State University will be dedicated to training the next generation of parallel programmers. 


About the PI

  Dr. Meilin Liu received her Ph. D. in Computer Science from The University of Texas at Dallas. She is an Associate Professor in the Department of Computer Science and Engineering at Wright State University. Her research interests include compiler optimization, program transformations, memory hierarchy optimization for multi-core systems, and GPU computing. Her research group is improving parallelism, and data locality by analyzing data dependencies and applying various program transformations or their combinations to improve the quality of code to take advantage of the hardware architecture of multi-core systems.

NVIDIA CUDA Center Blog announcement.