Automatic grain size determination for a macro dataflow real-time system

A compile time method for partitioning a periodic task into concurrent modules, called grains, and scheduling those modules for the shortest execution time possible is described. A suitable macro-dataflow real-time computer model with an improved instruction set (additional instructions for low-level synchronization have been introduced) executes a periodic task. This task must be first transformed from the DFCL (a relatively low-level dataflow real-time language) into a fine-grained dataflow graph, which is transformed next into a program graph. The compile time partitioning algorithm partitions a program graph into grains, and schedules them on the basis of the latest possible execution time without increasing the execution time of the whole program graph.<
>