@conference {1085, title = {Optimizing the SVD Bidiagonalization Process for a Batch of Small Matrices}, booktitle = {International Conference on Computational Science (ICCS 2017)}, year = {2017}, month = {2017-06}, publisher = {Procedia Computer Science}, organization = {Procedia Computer Science}, address = {Zurich, Switzerland}, abstract = {A challenging class of problems arising in many GPU applications, called batched problems, involves linear algebra operations on many small-sized matrices. We designed batched BLAS (Basic Linear Algebra Subroutines) routines, and in particular the Level-2 BLAS GEMV and the Level-3 BLAS GEMM routines, to solve them. We proposed device functions and big-tile settings in our batched BLAS design. We adopted auto-tuning to optimize different instances of GEMV routines. We illustrated our batched BLAS approach to optimize batched bi-diagonalization progressively on a K40c GPU. The optimization techniques in this paper are applicable to the other two-sided factorizations as well.}, doi = {https://doi.org/10.1016/j.procs.2017.05.237}, url = {http://www.sciencedirect.com/science/article/pii/S1877050917308645}, author = {Tingxing Dong and Azzam Haidar and Stanimire Tomov and Jack Dongarra} }