Title
Automatic code generation and tuning for stencil kernels on modern shared memory architectures
Abstract
In this paper, we present Patus, a code generation and auto-tuning framework for stencil computations targeted at multi- and manycore processors, such as multicore CPUs and graphics processing units. Patus, which stands for "Parallel Autotuned Stencils," generates a compute kernel from a specification of the stencil operation and a strategy which describes the parallelization and optimization to be applied, and leverages the autotuning methodology to optimize strategy-specific parameters for the given hardware architecture.
Year
DOI
Venue
2011
10.1007/s00450-011-0160-6
Computer Science - R&D
Keywords
Field
DocType
parallel autotuned stencils,stencil computations · code generation · autotuning · high performance computing,automatic code generation,modern shared memory architecture,present patus,stencil kernel,autotuning methodology,stencil computation,multicore cpus,manycore processor,hardware architecture,code generation,auto-tuning framework,stencil operation,shared memory,high performance computing
Shared memory,Supercomputer,Computer science,CUDA,Parallel computing,Stencil,Code generation,Compute kernel,Multi-core processor,Hardware architecture
Journal
Volume
Issue
ISSN
26
3-4
1865-2042
Citations 
PageRank 
References 
10
0.66
12
Authors
3
Name
Order
Citations
PageRank
Matthias Christen119010.20
Olaf Schenk253639.02
Helmar Burkhart330442.97