CG S, a new Language for Data-Parallel GPU Programming

Short Description
There are two promising GPU programming lan-. guages, C … Programming GPUs in assembler is tedious work. and error prone as well. With the release of Di- …

Website: rw4.cs.uni-sb.de | Filesize: 78kb

Content
CGIS, a new Language for Data-Parallel GPU Programming
Nicolas Fritz.. , Philipp Lucas.. , Philipp Slusallek
Universit? des Saarlandes, Germany
[cage|phlucas|slusallek]@cs.uni-sb.de
Abstract
In the last few years, GPUs have become new,
promising targets for general purpose programming.
Their inherent parallel architecture makes
them particularly suited for scientific numerical
computations with high arithmetical density.
There have been several proposals to exploit the
computational power of GPUs for data-parallel algorithms.
These approaches vary greatly in the abstraction
level of the graphics processing unit exposed
to the programmer.
Despite the emergence of GPU programming
languages, there is still need for a single high-level
programming language that naturally lends itself to
compilation into efficient code, yet does not overburden
the programmer with peculiarities of GPUs.
We present a novel unifying approach to facilitate
the use of GPUs for data-parallel computation.
We describe the language CGIS and the associated
infrastructure that allows scientific programmers to
express data-parallel computations efficiently on an
appropriate level of abstraction.
CR Descriptors: ACM Subject Classification: C.1.4:
Processor Architectures-Parallel Architectures; D.3.0:
Programming Languages-General; I.3.1: Computer
Graphics-Hardware Architecture
Keywords: GPUs, general purpose,…