Publication: Research - peer-review › Article in proceedings – Annual report year: 2012
In this paper, we consider the automatic performance tuning of dense vector and matrix-vector operations on GPUs. Such operations form the backbone of level 1 and level 2 routines in the Basic Linear Algebra Subroutines (BLAS) library and are therefore of great importance in many scientific applications. As examples, we develop single-precision CUDA kernels for the Euclidian norm (SNRM2) and the matrix-vector multiplication (SGEMV). The target hardware is the most recent Nvidia Tesla 20-series (Fermi architecture). We show that auto-tuning can be successfully applied to achieve high performance for dense vector and matrix-vector operations by appropriately utilizing the fine-grained parallelism of the GPU. Our tuned kernels display between 25-100% better performance than the current CUBLAS 3.2 library.
|Title of host publication||Parallel Processing and Applied Mathematics : 9th International Conference, PPAM 2011|
|Editors||Roman Wyrzykowski, Jack Dongarra, Konrad Karczewski, Jerzy Wasniewski|
|Conference||Parallel Processing and Applied Mathematics. 9th International Conference, PPAM 2011|
|Period||11/09/11 → 14/09/11|
|Name||Lecture Notes in Computer Science|
|Citations||Web of Science® Times Cited: No match on DOI|
- GPU, BLAS, Dense linear algebra, Parallel algorithms
Loading map data...