Automatic code generation of overlapped communications in a parallelisation tool

This paper addresses the exploitation of overlapping communication with calculation within parallel FORTRAN 77 codes for computational fluid dynamics (CFD) and computational structured dynamics (CSD). The obvious objective is to overlap interprocessor communication with calculation on each processor in a distributed memory parallel system and so improve the efficiency of the parallel implementation. A general strategy for converting synchronous to overlapped communication is presented together with tools to enable its automatic implementation in FORTRAN 77 codes. This strategy is then implemented within the parallelisation toolkit, CAPTools, to facilitate the automatic generation of parallel code with overlapped communications. The success of these tools are demonstrated on two codes from the NAS-PAR and PERFECT benchmark suites. In each case, the tools produce parallel code with overlapped communications which is as good as that which could be generated manually. The parallel performance of the codes also improve in line with expectation.

Multi-physics-multi-scale simulation and optimisation: the next generation

FEA and CFD analysis is becoming ever more complex with an emerging demand for simulation software technologies that can address ranges of problems that involve combinations of interactions amongst varying physical phenomena over a variety of time and length scales. Computation modelling of such problems requires software technologies that enable the representation of these complex suites of 'physical' interactions. This functionality requires the structuring of simulation modules for specific physical phemonmena so that the coupling can be effectiely represented. These 'multi-physics' and 'multi-scale' computations are very compute intensive and so the simulation software must operate effectively in parallel if it is to be used in this context. Of course the objective of 'multi-physics' and 'multi-scale' simulation is the optimal design of engineered systems so optimistation is an important feature of such classes of simulation. In this presentation, a multi-disciplinary approach to simulation based optimisation is described with some key examples of application to challenging engineering problems.

Computational modelling of multi-physics and multi-scale processes in parallel

This paper provides an overview of the developing needs for simulation software technologies for the computational modelling of problems that involve combinations of interactions amongst varying physical phenomena over a variety of time and space scales. Computational modelling of such problems requires software tech1nologies that enable the mathematical description of the interacting physical phenomena together with the solution of the resulting suites of equations in a numerically consistent and compatible manner. This functionality requires the structuring of simulation modules for specific physical phenomena so that the coupling can be effectively represented. These multi-physics and multi-scale computations are very compute intensive and the simulation software must operate effectively in parallel if it is to be used in this context. An approach to these classes of multi-disciplinary simulation in parallel is described, with some key examples of application to2 challenging engineering problems.

Multilevel refinement strategies for the capacity vehicle routing problem

We discuss the application of the multilevel (ML) refinement technique to the Vehicle Routing Problem (VRP), and compare it to its single-level (SL) counterpart. Multilevel refinement recursively coarsens to create a hierarchy of approximations to the problem and refines at each level. A SL algorithm, which uses a combination of standard VRP heuristics, is developed first to solve instances of the VRP. A ML version, which extends the global view of these heuristics, is then created, using variants of the construction and improvement heuristics at each level. Finally some multilevel enhancements are developed. Experimentation is used to find suitable parameter settings and the final version is tested on two well-known VRP benchmark suites. Results comparing both SL and ML algorithms are presented.