Home

An Embedded Language for Accelerated Array Computations

Data.Array.Accelerate defines an embedded language of array computations for high-performance computing in Haskell. Computations on regular, multi-dimensional arrays are expressed in the form of parameterised collective operations such as maps, reductions, and permutations. These computations are online-compiled and executed on a range of architectures.

• Quick Tour
• Availability
• Installation
• Importing the library
• Arrays, shapes, and indices
• Embedded arrays
• A stratified language
• Example: dot product
• Shapes
• Arrays
• Array types
• Getting data in
  • Lists
  • Data.Array.IArray
  • Data.Array.Repa
  • Data.Vector.Storable
  • Low-level conversions
  • Impure arrays
• The Accelerate language
• Troubleshooting
• Importing and running
• Developer's Guide
• Release Procedure

Benchmarks

• CUDA
• Dot product
• Black-Scholes option pricing
• N-Body simulation

External resources

• Simon Marlow's book Parallel and Concurrent Programming in Haskell contains a chapter on Accelerate.

• Simon Marlow recently taught a course on parallel & concurrent programming in Haskell at the EDF/CEA/INRIA Summer School 2012, "Functional Programming for Parallel and Concurrent Applications". One of the lectures teaches GPU programming using Accelerate; check out his [slides](http://community.haskell.org/~simonmar/slides/cadarache2012/7 - accelerate.pdf) and accompanying exercises (page 14 onwards).

• Ryan Newton describes the meta-par library, and how to use Accelerate as part of a parallel program that uses both the CPU and GPU. See his blog post here.