We report on the development of a visual programming environment for building applications for execution on a range of parallel computing platforms. This work exploits the dataflow and list-processing parallelism naturally exposed in the Prograph language, by providing annotations to indicate that operations can be remotely executed, supported by a task-pooling model for parallel execution that preserves the sequential semantics of the language. The goal is a practical system that builds on the comprehensive tools in the Prograph CPX environment in a consistent manner. This will be achieved via high-level editors for hardware configurations, annotations and program analysis, and an underlying kernel that implements the pooling model.
Philip T. Cox, Hugh Glaser, Stuart Maclean