%T Self\-Adjusting Mapping: A Heuristic Mapping Algorithm for Mapping Parallel Programs onto Transputer Networks
%A  Hong Shen
%E  J. Wexler
%B OUG\-11: Developing Transputer Applications
%X The problem of mapping parallel programs onto multiprocessor
   system is a fundmental problem of great significance in
   parallel processing, but it is NP\-hard in general. In this
   paper we propose a fast heuristic algorithm to solve this
   problem on transputer networks. Our mapping algorithm
   consists of three modules: grouping, placement and routing,
   where grouping groups processes in the program into tasks
   which can be placed onto processors in the transputer
   network in a way of one\-to\-one, placement places the
   grouped tasks onto the processors and routing produces
   physical communication paths for logical communication
   requirements. The three modules work co\-operatively in a
   way of progressive self\-adjusting, and finally produce a
   satisfactory solution for the mapping problem.