AUTOMATIC PARALLEL CODE GENERATION FOR MESSAGE-PASSING ON DISTRIBUTED-MEMORY SYSTEMS

The availability of a very accurate dependence graph for a scalar code is the basis for the automatic generation of an efficient parallel im plementation. The strategy for this task which is encapsulated in a co mprehensive data partitioning code generation algorithm is described. This algorithm involves the data partition, calculation of assignment ranges for partitioned arrays, addition of a comprehensive set of exec ution control masks, altering loop limits, addition and optimisation o f communications for all data. In this context, the development and im plementation of strategies to merge communications wherever possible h as proved an important feature in producing efficient parallel impleme ntations for numerical mesh based codes. The code generation strategie s described here are embedded within the Computer Aided Parallelisatio n tools (CAPTools) software as a key part of a toolkit for automating as much as possible of the parallelisation process for mesh based nume rical codes. The algorithms used enables parallelisation of real compu tational mechanics codes with only minor user interaction and without any prior manual customisation of the serial code to suit the parallel isation tool.