Section: New Results
The last two years we studied several heuristics which are extensions of the one used in AAA/SynDEx tolerating a fixed number of arbitrary processors, links (point-to-point communication medium), and buses (multipoint communication medium) faults. Moreover, we proposed a heuristic generating reliable distributions and schedulings and taking into account two criteria: the minimization of the latency (execution duration of the distributed and scheduled algorithm onto the architecture) and the maximization of the reliability of the processors and the communication media. These heuristics are all based on automatic software redundancy. They have been detailed in  and implemented in the version 6 of SynDEx.