The development of distributed—memory muItiprocessing system(DMS)needs the support of the corresponding parallelization software tool．In current paralleliza-tion systems，the user must explicitly specify how the data domain of the sequential program is to be partitioned and mapped to the processors of DMS．In this paper，the authors present a knowledge based data automatic distribution model(KBDM)that provides au-tomatic support for this task，and then discuss the principal ideas underlying their model，the major components of KBDM and the main features of the distribution strategy employed．