NVIDIA announced this morning that they have recognized the University of Tennessee, Knoxville’s [UTK] Innovative Computing Laboratory [ICL] as a CUDA Center of Excellence. They made special note of UTK/ICL’s adoption of the CUDA programming model into its curriculum as well as its work in developing linear algebra libraries for use in HPC. UTK joins a select group of seven other universities and research organizations that include Harvard, Cambridge and the National Taiwan University.
This award of a CUDA Center of Excellence underscores ICL’s commitment to continue our work at the forefront of high performance, scientific computing,” said Jack Dongarra, ICL’s director. “We are very proud of this award and excited by the opportunity it affords to pursue our research on NVIDIA’s groundbreaking platform.”
NVIDIA technologies are now well established in the forefront of general purpose, parallel computing. Our work on the development of Linear Algebra Libraries for CUDA-based Hybrid Architectures will further enable and expand these technologies in the general area of high-performance scientific computing. MAGMA, a subset of LAPACK for CUDA-based Hybrid Architectures, is only a first step in this direction,” added Dongarra.
Mathematical algorithms and their respective [efficient] implementations are key to getting quite a few applications to really fly. For more info, check out the CUDA Center of Excellence page here.