Performance Optimization and Modeling of Blocked Sparse Kernels

13 years 5 months ago

Download www.cs.utk.edu

We present a method for automatically selecting optimal implementations of sparse matrixvector operations. Our software ‘AcCELS’ (Accelerated Compress-storage Elements for Linear Solvers) involves a setup phase that probes machine characteristics, and a run-time phase where stored characteristics are combined with a measure of the actual sparse matrix to ﬁnd the optimal kernel implementation. We present a performance model that is shown to be accurate over a large range of matrices.

Alfredo Buttari, Victor Eijkhout, Julien Langou, S

Real-time Traffic

Actual Sparse Matrix | IJHPCA 2007 | Optimal Kernel Implementation | Sparse Matrixvector Operations |

claim paper

» Exploring the effect of block shapes on the performance of sparse kernels

» Optimizing Sparse Matrix Computations for Register Reuse in SPARSITY

» A Comparative Study of Blocking Storage Methods for Sparse Matrices on Multicore Architect...

» Sparse Gaussian graphical models with unknown block structure

» NearOracle Performance of Greedy BlockSparse Estimation Techniques from Noisy Measurements

» A Family of HighPerformance Matrix Multiplication Algorithms

» Support Kernel Machines for Object Recognition

» MinimaxOptimal Rates For Sparse Additive Models Over Kernel Classes Via Convex Programming

Post Info
More Details (n/a)

Added	15 Dec 2010
Updated	15 Dec 2010
Type	Journal
Year	2007
Where	IJHPCA
Authors	Alfredo Buttari, Victor Eijkhout, Julien Langou, Salvatore Filippone

Comments (0)

Sciweavers

Performance Optimization and Modeling of Blocked Sparse Kernels

Actual Sparse Matrix | IJHPCA 2007 | Optimal Kernel Implementation | Sparse Matrixvector Operations |

Explore & Download

Productivity Tools

Document Tools

Image Tools

Sciweavers