Dr. Mustafa KH. Aswad

I am recently got  my  PhD  in  Computer Science  from  School of Mathematics and Computer Science at Heriot-Watt University.

Research

My research interest  focuses on high-level programming models for parallel computation. In particular, I am working on the implementation of Glasgow parallel Haskell (GpH). GpH is  a distributed parallel implantation of Glasgow Haskell Compiler.  I have extend the GpH implementation with architecture-aware mechanism deals with memory and core hierarchy of  multicore clusters .  

Useful  Tools

[Darcs

, GpH, Latex]

Publications

Architecture Aware Parallel Programming in Glasgow Parallel Haskell (GPH)

M. KH. Aswad, P. W. Trinder,  H. W. {Loidl},  Accepted in   to 9th International Workshop on Practical Aspects of High-Level Parallel Programming (PAPP 2012)

June 4-6, 2012 – – Omaha, USA

Seq no more: Better Strategies for Parallel Haskell

              Simon Marlow, Patrick Maier, Hans-Wolfgang Loidl, Mustafa K. Aswad, and Phil Trinder.

              Haskell Symposium 2010, Baltimore, MD, United States, 30th September, 2010

 Low Pain vs No Pain Multi-core Haskells

M. KH.  Aswad, P. W. Trinder, A.D.AL Zain, G.J. Michealson and Jost Berthold, The 10th Symposium on trends in functional programming TFP 2009 Selye Janos University, Komarno, Slovakia, June 2-4, 2009.

 

Low Pain vs No Pain Multi-core Haskells (Full report)

M. KH.  Aswad, P. W. Trinder, A.D.AL Zain, G.J. Michealson and Jost Berthold, Technical Report No: 0068 , School of Mathematics and Computer Sciences Heriot-Watt University,Edinburgh, UK.

Low-Pain, High-Gain Multicore Programming in Haskell: Coordinating Irregular Symbolic Computations on MultiCore Architectures.

 ACM SIGPLAN Workshop on Declarative Aspects of Multicore Programming (DAMP'09), Savannah, Georgia, USA (January 2009).

Multi-architecture Parallel Programming using GpH, a Functional Language

MPhil thesis, School of Mathematical and Computer Sciences, Heriot-Watt University, January 2003.

Contact Details

Department of Computer Science,
School of Mathematical and Computer Sciences,
Heriot-Watt University,
Riccarton,
Edinburgh EH14 4AS
Scotland, U.K.

Room: G59
Tel.: +44 131 451 4183
Fax: +44 131 451 3431
email: mka19@hw.ac.uk


 

 

Abstuct:

The new multi-core platforms which become available on our  desk with  large numbers of processors in one chip  open  a big challenge to the developers of parallel programming models. They are required to provide a model which has multi levels  of parallel  abstractions. The different levels of abstraction of parallel processors may be in multi-core,  multiprocessor or between multiple computers.  A high level parallel programming languages is good for achieving different levels of abstraction with an acceptable performance across a wide rang of parallel platforms.  Functional  programming languages which fulfil the above criteria, are theoretically well suited to exploiting variety of  multi-core platforms. However, implementing effective high level parallel functional languages is still requires  lots  effort to improve the efficiency.   Therefore  I am plan  to produce a parallel programming model which will be a shared  and distributed running  on grid of non-uniform Processor Hierarchies platforms