Learn More
One central problem in the execution of parallel nested loops with non-aane bounds is the precise scanning (i.e., enumeration) of the points in their iteration space and the detection of their termination. Scanning schemes have been proposed for both shared-memory and distributed-memory implementations. However, these schemes work only for perfectly nested(More)
This thesis deals with the automatic parallelization step of compilers for multi-processor computers. The focus is on practical methods for this task, which means methods that are general and fast. Thus, a wide variety of loop programs is accepted as input, including imperfectly nested loops with aane variable accesses and dependences. The scheduler assigns(More)
  • 1