Hing Wing To

Parallel skeletons for structured composition

In this paper, we propose a straightforward solution to the problems of compositional parallel programming by using skeletons as the uniform mechanism for structured composition. In our approach parallel programs are constructed by composing procedures in a conventional base language using a set of high-level, pre-defined, functional, parallel computational forms known as skeletons. The ability to compose skeletons provides us with the essential tools for building further and more complex application-oriented skeletons specifying important aspects of parallel computation. Compared with the process network based composition approach, such as PCN, the skeleton approach abstracts away the fine details of connecting communication ports to the higher level mechanism of making data distributions conform, thus avoiding the complexity of using lower level ports as the means of interaction. Thus, the framework provides a natural integration of the compositional programming approach with the data parallel programming paradigm.

Functional Skeletons for Parallel Coordination

In this paper we propose a methodology for structured parallel programming using functional skeletons to compose and coordinate concurrent activities written in a standard imperative language. Skeletons are higher order functional forms with built-in parallel behaviour. We show how such forms can be used uniformly to abstract all aspects of a parallel programs behaviour including data partitioning, placement and re-arrangement (communication) as well as computation. Skeletons are naturally data parallel and are capable of expressing computation and co-ordination at a higher level of abstraction than other process oriented co-ordination notations. Examples of the application of this methodology are given and an implementation technique outlined.

Structured parallel programming

Parallel programming is a difficult task involving many complex issues such as resource allocation, and process coordination. We propose a solution to this problem based on the use of a repertoire of parallel algorithmic forms, known as skeletons. The use of skeletons enables the meaning of a parallel program to be separated from its behaviour. Central to this methodology is the use of transformations and performance models. Transformations provide portability and implementation choices, whilst performance models guide the choices by providing predictions of execution time. We describe the methodology and investigate the use and construction of performance models by studying an example.<<ETX>>

Co-ordinating Heterogeneous Parallel Computation

There is a growing interest in heterogeneous high performance computing environments. These systems are difficult to program owing to the complexity of choosing the appropriate resource allocations and the difficulties in expressing these choices in traditional parallel languages. In this paper we propose that functional skeletons are used to express these resource allocation strategies. By associating performance models with each skeleton it is possible to predict and optimise the performance of different resource allocation strategies, thus providing a tool for guiding the choice of resource allocation. Through a case study of a parallel conjugate gradient algorithm on a mixed vector and scalar parallel machine we demonstrate these features of the SPP(X) approach.

Parallel Induction Algorithms for Data Mining

In the last decade, there has been an explosive growth in the generation and collection of data. Nonetheless, the quality of information inferred from this voluminous data has not been proportional to its size. One of the reasons for this is that the computational complexities of the algorithms used to extract information from the data are normally proportional to the number of input data items resulting in prohibitive execution time on large data sets. Parallelism is one solution to this problem. In this paper we present preliminary results on experiments in parallelising C4.5, a classification-rule learning system using decision-trees as a model representation, which has been used as a base model for investigating methods for parallelising induction algorithms. The experiments assess the potential for improving the execution time by exploiting parallelism in the algorithm.