Section: Software

Implementations of Synchronous Programs

Participant : Alain Girault.

Fault Tolerance

We have been cooperating for several years with the Inria team Aoste (Inria Sophia-Antipolis and Rocquencourt) on the topic of fault tolerance and reliability of safety critical embedded systems. In particular, we have implemented several new heuristics for fault tolerance and reliability within their software SynDEx (http://www-rocq.inria.fr/syndex ). Our first scheduling heuristic produces static multiprocessor schedules tolerant to a specified number of processor and communication link failures [55] . The basic principles upon which we rely to make the schedules fault tolerant is, on the one hand, the active replication of the operations [56] , and on the other hand, the active replication of communications for point-to-point communication links, or their passive replication coupled with data fragmentation for multi-point communication media (i.e., buses) [57] . Our second scheduling heuristic is multi-criteria: it produces a static schedule multiprocessor schedule such that the reliability is maximized, the power consumption is minimized, and the execution time is minimized [3] [33] [17] , [11] . Our results on fault tolerance are summarized in a web page (http://pop-art.inrialpes.fr/~girault/Projets/FT ).