Schedule Based Code Generation for ParallelProcessors

Abstract: Dynamic model driven architecture (DMDA) is a architecture made to aid in the development of parallel computing code. This thesis is applied to an implementation of DMDA known as DMDA3 that should convert graphs of computations into efficient computation code, and it deals with the translation of Platform Specific Models (PSM) into running systems. Currently DMDA3 can generate schedules of operations but not finished code. This thesis describes a DMDA3 module that turns a schedule of operations into a runable program. Code was obtained from the DMDA3 schedules by reflection and a framework was build that allowed generation of low level language code from schedules. The module is written in Java and can currently generate C and Fortran code for computational tasks. Based on runtime tests for matrix multiplication algorithms the generated code is almost as fast as handwritten code.