Data Layout Transformation for Stencil Computations on Short-Vector SIMD Architectures
| Autor: | Thomas Henretty, Louis-Noël Pouchet, P. Sadayappan, Franz Franchetti, J. Ramanujam, Kevin Stock |
|---|---|
| Rok vydání: | 2011 |
| Předmět: |
Stencil code
Computer science 90699 Electrical and Electronic Engineering not elsewhere classified Parallel computing Static analysis ComputerSystemsOrganization_PROCESSORARCHITECTURES Stencil Single-precision floating-point format Instruction set Transformation (function) FOS: Electrical engineering electronic engineering information engineering Computational electromagnetics SIMD Computer Engineering |
| Zdroj: | Lecture Notes in Computer Science ISBN: 9783642198601 CC |
| DOI: | 10.1184/r1/6468599 |
| Popis: | Stencil computations are at the core of applications in many domains such as computational electromagnetics, image processing, and partial differential equation solvers used in a variety of scientific and engineering applications. Short-vector SIMD instruction sets such as SSE and VMX provide a promising and widely available avenue for enhancing performance on modern processors. However a fundamental memory stream alignment issue limits achieved performance with stencil computations on modern short SIMD architectures. In this paper, we propose a novel data layout transformation that avoids the stream alignment conflict, along with a static analysis technique for determining where this transformation is applicable. Significant performance increases are demonstrated for a variety of stencil codes on three modern SIMD-capable processors. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|