Nearly two dozen more universities and research institutions were added this past quarter to our roster of CUDA Research Centers and CUDA Teaching Centers, bringing the total to 229.
The work being done at the 22 new CUDA centers in eight countries includes sparse numerical linear algebra, heterogeneous computing, astrophysical fluid dynamics and next generation DNA sequencing. There are now CUDA Research and Teaching centers in 41 countries.
CUDA Teaching Centers equip tens of thousands of students graduating each year with the knowledge and expertise to take advantage of the parallel processing power of GPUs (see “What Is CUDA?”). These centers get free teaching kits, textbooks, software licenses, NVIDIA CUDA architecture-enabled GPUs for teaching lab computers and academic discounts for additional hardware.
CUDA Research Centers embrace GPU computing across multiple research fields. They have access to exclusive events with key researchers and academics, a designated NVIDIA technical liaison and specialized training sessions.
Here are some examples of CUDA-related work taking place at some of our newest CUDA Research Centers:
King Abdullah University of Science and Technology (Saudi Arabia)
The research agenda of the new CUDA Research Center at KAUST is driven by three challenging problems. The first problem is with the implementation of efficient dense/sparse numerical linear algebra kernels (the KAUST-BLAS, or KBLAS, library) for direct and iterative solvers on heterogeneous architectures. The second is designing Fast Multipole Methods (the exaFMM library) to solve some of the critical high performance computing problems applications in fluid dynamics, molecular dynamics, and pre-conditioners for sparse linear solvers. The third deals with extreme-scale visualization through novel algorithms for research areas including neuroscience, seismic exploration, materials science, mechanical engineering and bioinformatics.
Rochester Institute of Technology (US)
RIT is currently working on acceleration of patient-specific modeling applications. Due to their heterogeneous nature and linear algebra complexity, these applications can make use of heterogeneous hardware solutions. In order to quickly explore the design space of these solutions, we intend to accurately model GPU platforms. They are also working on more efficient usage of the memory structures of the GPU.
Technical University of Denmark (Denmark)
The research team of GPUlab focuses on developing efficient algorithms and software components based on modern parallel programming paradigms for heterogeneous computing. using many-core hardware such as GPUs.
Other new CUDA Research Centers include:
• Indian Institute of Science, Bangalore (India)
• La Maison de la Simulation (France)
• Norwegian University of Science & Technology (Norway)
• University of Heidelberg (Germany)
• University of Uppsala (Sweden)
The new CUDA Teaching Centers include:
• Chitkara University (India)
• La Maison de la Simulation (France)
• Lawrence Technological University (U.S.)
• Trinity College (US)
• Universidad Politécnica de Madrid (Spain)
• Universidade do Vale do Itajaí (Brazil)
• Universidade Federal de Goiás (Brazil)
• Universitat d’Alicant (Spain)
• Universität des Saarlandes (Germany)
• University of Dallas (U.S.)
• University of Heidelberg (Germany)
• University of Surrey (UK)
• VŠB-Technical University of Ostrava (Czech Republic)
• Walchand College of Engineering (India)
For more information on NVIDIA research activities and these programs, please visit the NVIDIA Research site.
Tagged: crc, ctc, CUDA, GPGPU, gpu, gpu computing, parallel computing
