This thesis proposes a novel model of computation, called stream rewriting, for the specification and implementation of highly concurrent applications. Basically, the active tasks of an application and their dependencies are encoded as a token stream, which is iteratively modified by a set of rewriting rules at runtime. In order to estimate the performance and scalability of stream rewriting, a large number of experiments have been evaluated on many-core systems and the task management has been implemented in software and hardware.
