GpH -- Glasgow Parallel Haskell

Activity Report November 2003

Report by: Phil Trinder

The Team

Phil Trinder, Kevin Hammond, Hans-Wolfgang Loidl, Abyd Al Zain, Jost Berthold, Xiao Yan Deng, Murray Gross, Andre Rauber du Bois, Alvaro Rebon Portillo, Leonid Timochouk.

Status

A complete, GHC-based implementation of the parallel Haskell extension GpH and of evaluation strategies is available. Extensions of the runtime-system and language modules, to improve performance and support for architecture-independence, are under development.

Implementations

The GUM implementation of GpH is available in two development branches.

The stable branch (GUM-4.06, based on GHC-4.06) is available for RedHat-based Linux machines: binary snapshot (see installation instructions). The stable branch is available from the GHC CVS repository via tag gum-4-06.
The unstable branch (GUM-5.02, based on GHC-5.02) is currently being tested on a Beowulf cluster. Most of our test programs run already, with minor issues left to be resolved before this version will become our main development version. The unstable branch is available from the GHC CVS repository via tag gum-5-02-3.

Our main hardware platform are Intel-based Beowulf clusters. Work on ports to other architectures is also moving on (and available on request):

A port to a Sun-Solaris shared-memory machine exists but currently suffers from performance problems, which we are trying to track down.
A port to a Mosix cluster has been built in the Metis project at Brooklyn College, with a first version available on request from Murray Gross.
The Eden version from the GHC CVS(tag eden-5-02-3) supports both GpH and Eden languages.

System Evaluation and Enhancement

We have ported GUM to computational GRIDs, replacing the current PVM layer with IMPICH-G2. We have measured programs on a single grid-enabled cluster, and are very close to measuring programs on more than 1 cluster.
We are teaching GpH to students at Heriot-Watt and Phillips Universitat Marburg.
We are designing a combined, modular Eden/GpH implementation.
We have compared the implementation designs of GpH and PMLS.

GpH Applications

We have outlined a methodology for developing parallel programs in an architecture independent fashion.

Language

We have constructed efficient implementations of several algorithmic skeletons in GdH, showing that they are more efficient, but less flexible than evaluation strategies.

3.3.2.1 Further reading:

GpH Home Page