Dominik Grewe
Anton Lokhmotov
ABSTRACT
We propose a system-independent representation of sparse matrix formats that allows a compiler to generate efficient, system-specific code for sparse matrix operations. To show the viability of such a representation we have developed a compiler that generates and tunes code for sparse matrix-vector multiplication (SpMV) on GPUs. We evaluate our framework on six state-of-the-art matrix formats and show that the generated code performs similar to or better than hand-optimized code.
- Nathan Bell and Michael Garland. Implementing sparse matrix-vector multiplication on throughput-oriented processors. In SC, 2009. Google Scholar
Digital Library
- Jee W. Choi, Amik Singh, and Richard W. Vuduc. Model-driven autotuning of sparse matrix-vector multiply on GPUs. In PPoPP, 2010. Google ScholarDigital Library
- David R. Kincaid, John R. Respess, and David M. Young. ITPACK 2.0 user's guide. Technical Report CNA-150, Center for Numerical Analysis, University of Texas, Austin, Texas, 1979.Google Scholar
- Nikolay Mateev, Keshav Pingali, Paul Stodghill, and Vladimir Kotlyar. Next-generation generic programming and its application to sparse matrix computations. In ICS, 2000. Google ScholarDigital Library
- Alexander Monakov, Anton Lokhmotov, and Arutyun Avetisyan. Automatically tuning sparse matrix-vector multiplication for GPU architectures. In HiPEAC, 2010. Google ScholarDigital Library
- Shane Ryoo, Christopher I. Rodrigues, Sara S. Baghsorkhi, Sam S. Stone, David B. Kirk, and Wen-mei W. Hwu. Optimization principles and application performance evaluation of a multithreaded GPU using CUDA. In PPoPP, 2008. Google ScholarDigital Library
- Richard W. Vuduc. Automatic performance tuning of sparse matrix kernels. PhD thesis, University of California, Berkeley, CA, USA, 2004. Google ScholarDigital Library
- Samuel Williams, Leonid Oliker, Richard W. Vuduc, John Shalf, Katherine A. Yelick, and James Demmel. Optimization of sparse matrix-vector multiplication on emerging multicore platforms. Parallel Computing, 2009. Google ScholarDigital Library
Index Terms
Automatically generating and tuning GPU code for sparse matrix-vector multiplication from a high-level representation
Recommendations
On Implementing Sparse Matrix Multi-vector Multiplication on GPUs
Sparse matrix-vector and multi-vector multiplications (SpMV and SpMM) are performance bottlenecks operations in numerous HPC applications. A variety of SpMV GPU kernels using different matrix storage formats have been developed to accelerate these ...
An Effective Approach for Implementing Sparse Matrix-Vector Multiplication on Graphics Processing Units
Sparse matrix vector multiplication, SpMV, is often a performance bottleneck in iterative solvers. Recently, Graphics Processing Units, GPUs, have been deployed to enhance the performance of this operation. We present a blocked version of the Transposed ...
A Portable and High-Performance General Matrix-Multiply (GEMM) Library for GPUs and Single-Chip CPU/GPU Systems
OpenCL is a vendor neutral and portable interface for programming parallel compute devices such as GPUs. Tuning OpenCL implementations of important library functions such as dense general matrix multiply (GEMM) for a particular device is a difficult ...
Comments