From shared to distributed memory: converting non-numeric parallel algorithms to Message Passing Interface (MPI) | IEEE Conference Publication | IEEE Xplore