This paper presents a practical evaluation and comparison of three stateof -the-art parallel functional languages. The evaluation is based on implementations of three typical symbolic computation programs, with performance measured on a Beowulf-class parallel architecture.

High-level control of parallel process behaviour simplifies the development of parallel software substantially by freeing the programmer from low-level process management and coordination details. The latter are handled by a sophisticated runtime system which controls program execution. In this paper we look behind the scenes and show how the enormous gap between high-level parallel language constructs and their low-level implementation has been bridged in the implementation of the parallel functional language Eden. The main idea has been to specify the process control in a functional language and to restrict the extensions of the low-level runtime system to a few selected primitive operations.

First of all, I would like to thank Prof. Dr. Claudia Leopold for giving me the opportunity to work in her group of “Programming Languages and Parallel Programming ” at University of Kassel and Prof. Dr. Emre Harmancı who accepted to be my supervisor. Prof. Leopold never let me alone even in the rest of my work in Turkey. Working at University of Kassel was a great experience for me. I feel fortunate to have helpful friends, Björn Knafla and Michael Süss who made my first time in Germany enjoyable. Thanks to Christiane Becker and Raffaele Biscosi for being so nice to me. I am grateful to Dr. Turgay Altılar who encouraged me for the presentations and helped me a lot. Special thanks to my parents who always supported me. December 2004 Ay¸se Beliz S¸enyüz iv TABLE OF CONTENTS ABBREVIATIONS viii

This paper presents Haskell# , a parallel functional language based on coordination. Haskell# supports lazy stream communication and facilities, at coordination level, to the specification of data parallel programs. Haskell# supports a clean and complete, semantic and syntactic, separation between coordination and computation levels of programming, with several benefits to parallel program engineering. The implementation of some well-known applications in Haskell# is presented, demonstrating its expressiveness, allowing for elegant, simple, and concise specification of any static pattern of parallel, concurrent or distributed computation.

In this paper, it is shown how skeletons can be introduced into Haskell # at configuration level, by extending its notion of hierarchical composition of programs with process templates. The approach described herein is general enough to be applied to configuration languages in general. Its expressiveness, simplicity and elegance are demonstrated by examples, which also show its impact in Haskell # programming practice and performance.

This paper presents Haskell#, a parallel functional language based on coordination. Haskell# supports lazy stream communication and facilities, at coordination level, to the specification of data parallel programs. Haskell# supports a clean and complete, semantic and syntactic, separation between coordination and computation levels of programming, with several benefits to parallel program engineering. The implementation of some well-known applications in Haskell# is presented, demonstrating its expressiveness, allowing for elegant, simple, and concise specification of any static pattern of parallel, concurrent or distributed computation.

In this paper, we present the parallelization of a sequential functional implementation of a Monte Carlo Transport Problem, called MCP- Haskell[Hammes et al., 1995], using Haskell# . This experiment gave us important feedback for evaluating Haskell# features, helping us to answer some questions, like how expressive is Haskell# for representing known parallel computational patterns, how easy it is to build large scale parallel programs in an elegant and concise way, and how efficient are Haskell# programs. Based on our conclusions, we suggest new features to be incorporated in Haskell# to improve its expressiveness and performance. We also present the performance figures for the MCP-Haskell# benchmark.