Automatically generating and tuning GPU code for sparse matrix-vector multiplication from a high-level representation

Authors:
Dominik Grewe;Anton Lokhmotov
Affiliations:
University of Edinburgh, UK;ARM, Cambridge, UK
Venue:
Proceedings of the Fourth Workshop on General Purpose Processing on Graphics Processing Units
Year:
2011

Citing 7
Cited 4

Next-generation generic programming and its application to sparse matrix computations

Proceedings of the 14th international conference on Supercomputing
Automatic performance tuning of sparse matrix kernels

Automatic performance tuning of sparse matrix kernels
Optimization principles and application performance evaluation of a multithreaded GPU using CUDA

Proceedings of the 13th ACM SIGPLAN Symposium on Principles and practice of parallel programming
Optimization of sparse matrix-vector multiplication on emerging multicore platforms

Parallel Computing
Implementing sparse matrix-vector multiplication on throughput-oriented processors

Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis
Model-driven autotuning of sparse matrix-vector multiply on GPUs

Proceedings of the 15th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming
Automatically tuning sparse matrix-vector multiplication for GPU architectures

HiPEAC'10 Proceedings of the 5th international conference on High Performance Embedded Architectures and Compilers

clSpMV: A Cross-Platform OpenCL SpMV Framework on GPUs

Proceedings of the 26th ACM international conference on Supercomputing
Input-aware auto-tuning for directive-based GPU programming

Proceedings of the 6th Workshop on General Purpose Processor Using Graphics Processing Units
SMAT: an input adaptive auto-tuner for sparse matrix-vector multiplication

Proceedings of the 34th ACM SIGPLAN conference on Programming language design and implementation
Accelerating sparse matrix-vector multiplication on GPUs using bit-representation-optimized schemes

SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

Quantified Score

Hi-index	0.00

Visualization

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.