Abstract It has often been suggested that functional languages provide an excellent basis for programming parallel computer systems. This is largely a result of the lack of side effects which makes it possible to evaluate the subexpressions of a given term without any risk of interference. On the other hand, the lack of side-effects has also been seen as a weakness of functional languages since it rules out many features of traditional imperative languages such as state, I/O and exceptions. These ideas can be simulated in a functional language but the resulting programs are sometimes unnatural and inefficient. On the bright side, recent work has shown how many of these features can be naturally incorporated into a functional language without compromising efficiency by expressing computations in terms of monads or continuations. Unfortunately, the "single-threading " implied by these techniques often destroys many opportunities for parallelism. In this paper, we describe a simple extension to the Haskell I/O monad that allows a form of explicit high-level concurrency. It is a simple matter to incorporate these features in a sequential implementation, and genuine parallelism can be obtained on a parallel machine. In addition, the inclusion of constructs for explicit concurrency enhances the use of Haskell as an executable specification language, since some programs are most naturally described as a composition of parallel processes. \Lambda This research was supported by ARPA via a subcontract to Intermetrics, Inc. 1

This work compares three algorithms for the three dimensional N-body problem, the Barnes-Hut algorithm, Greengard's Fast Multipole Method(FMM), and the Parallel Multipole Tree Algorithm (PMTA) to determine which of the algorithms performs best in practice. Although FMM has a better asymptotic running time (O(N ) instead of O(N log N ) for uniform distributions), the algorithm is more complicated and it is not immediately clear above what values of N it performs better in practice. We studied the dependence of accuracy on the variable parameters `, p and ff, and then compared the floating point operation counts of the three algorithms at similar levels of accuracy, for both charged and uncharged random distributions. At a high level of accuracy (RMS-error ß 10 \Gamma5 ), the FMM did the least number of operations for N ? 10 4 , assuming both charged and uncharged distributions of points. At a lower level of accuracy, (RMS-error ß 10 \Gamma3 ) for uncharged distributions, the FMM d...

Given an ensemble of n bodies in space whose interaction is governed by a potential function, the N-body problem is to calculate the force on each body in the ensemble that results from its interaction with all other bodies. An efficient algorithm for this problem is critical in the simulation of molecular dynamics, turbulent fluid flow, intergalactic matter and other problems. The fast multipole algorithm (FMA) developed by Greengard approximates the solution with bounded error in time O(n ). For non-uniform distributions of bodies, an adaptive variation of the algorithm is required to maintain this time complexity. The parallel execution of the FMA poses complex implementation issues in the decomposition of the problem over processors to reduce communication. As a result the 3D Adaptive FMA has, to our knowledge, never been implemented on a scalable parallel computer. This paper describes several variations on the parallel adaptive 3D FMA algorithm that are expressed using the datapa...

A prototyping approach to the implementation of complex parallel algorithms on actual parallel machines is proposed as a way to bridge the gap between theoretical descriptions of parallel algorithms and the implementationlevel details required in current low-level parallel programming languages. Prototypes are constructed using the wide-spectrum parallel programming language Proteus and can be executed using a Proteus interpreter. The widespectrum nature of the Proteus language permits key aspects of the implementation to be carefully investigated while other, ancillary, aspects can remain at a convenient level of abstraction. Manual and automatic techniques are under investigation to transform prototypes to low-level programming languages for specific machines. In this paper several different N-body force calculation techniques are used to introduce Proteus, including a preliminary implementation study of the 3-dimensional fast multipole algorithm of Greengard, an algorithm of current...