A TRANSPUTER-BASED PARALLEL MACHINE TRANSLATION SYSTEM FOR INDIAN LANGUAGES

This paper presents the performance studies carried out in the design of a parallel machine translation system (MTS). It has been implemente d in the language Occam-2 on a PC/AT hardware platform augmented with a 3-transputer add-on card. Exploitation of parallelism and a good dic tionary organization are identified as the two ways of providing an ef ficient solution. Parallelism in the algorithm for the morphological a nd phrase-level analyses and in the generation phases of translation h ave been fully utilized. The performance of the MTS is discussed for v arious methods of task distribution and the way the dictionary is acce ssed. The performance studies have also been conducted on the MTS with and without load sharing among the transputers. The results show that the system performance is nearly doubled with the addition of the tra nsputer card and with the transputers working under no-load sharing co ndition. (C) 1997 Elsevier Science B.V.