Efficient parallel stencil convolution in Haskell

Authors:
Ben Lippmeier;Gabriele Keller
Affiliations:
University of New South Wales, Sydney, Australia;University of New South Wales, Sydney, Australia
Venue:
Proceedings of the 4th ACM symposium on Haskell
Year:
2011

Citing 22
Cited 5

Detecting equality of variables in programs

POPL '88 Proceedings of the 15th ACM SIGPLAN-SIGACT symposium on Principles of programming languages
Global value numbers and redundant computations

POPL '88 Proceedings of the 15th ACM SIGPLAN-SIGACT symposium on Principles of programming languages
Lazy functional state threads

PLDI '94 Proceedings of the ACM SIGPLAN 1994 conference on Programming language design and implementation
Computer aided parallelisation tools (CAPTools)—conceptual overview and performance on the parallelisation of structured mesh codes

Parallel Computing
ZPL: A Machine Independent Programming Language for Parallel Computers

IEEE Transactions on Software Engineering - Special issue on architecture-independent languages and software tools for parallel processing
Improving Software Pipelining With Unroll-and-Jam

HICSS '96 Proceedings of the 29th Hawaii International Conference on System Sciences Volume 1: Software Technology and Architecture
Finding Edges and Lines in Images

Finding Edges and Lines in Images
Titanium Language Reference Manual

Titanium Language Reference Manual
Single Assignment C: efficient support for high-level array operations in a functional setting

Journal of Functional Programming
Effective automatic parallelization of stencil computations

Proceedings of the 2007 ACM SIGPLAN conference on Programming language design and implementation
Stream fusion: from lists to streams to nothing at all

ICFP '07 Proceedings of the 12th ACM SIGPLAN international conference on Functional programming
Practical Algorithms for Image Analysis with CD-ROM

Practical Algorithms for Image Analysis with CD-ROM
Stencil computation optimization and auto-tuning on state-of-the-art multicore architectures

Proceedings of the 2008 ACM/IEEE conference on Supercomputing
A type and effect system for deterministic parallel Java

Proceedings of the 24th ACM SIGPLAN conference on Object oriented programming systems languages and applications
PASTHA: parallelizing stencil calculations in Haskell

Proceedings of the 5th ACM SIGPLAN workshop on Declarative aspects of multicore programming
Ypnos: declarative, parallel structured grid programming

Proceedings of the 5th ACM SIGPLAN workshop on Declarative aspects of multicore programming
The computational energy spectrum of a program as it executes

The Journal of Supercomputing
An llVM backend for GHC

Proceedings of the third ACM Haskell symposium on Haskell
Supercompilation by evaluation

Proceedings of the third ACM Haskell symposium on Haskell
Regular, shape-polymorphic, parallel arrays in Haskell

Proceedings of the 15th ACM SIGPLAN international conference on Functional programming
Rethinking supercompilation

Proceedings of the 15th ACM SIGPLAN international conference on Functional programming
Accelerating Haskell array codes with multicore GPUs

Proceedings of the sixth workshop on Declarative aspects of multicore programming

Guiding parallel array fusion with indexed types

Proceedings of the 2012 Haskell Symposium
Optimising purely functional GPU programs

Proceedings of the 18th ACM SIGPLAN international conference on Functional programming
Exploiting vector instructions with generalized stream fusio

Proceedings of the 18th ACM SIGPLAN international conference on Functional programming
Automatic SIMD vectorization for Haskell

Proceedings of the 18th ACM SIGPLAN international conference on Functional programming
An EDSL approach to high performance Haskell programming

Proceedings of the 2013 ACM SIGPLAN symposium on Haskell

Quantified Score

Hi-index	0.00

Visualization

Abstract

Stencil convolution is a fundamental building block of many scientific and image processing algorithms. We present a declarative approach to writing such convolutions in Haskell that is both efficient at runtime and implicitly parallel. To achieve this we extend our prior work on the Repa array library with two new features: partitioned and cursored arrays. Combined with careful management of the interaction between GHC and its back-end code generator LLVM, we achieve performance comparable to the standard OpenCV library.