A. Girault, C. Lavarenne, M. Sighireanu, and Y. Sorel. Faulttolerant static scheduling for real-time distributed embedded systems. Research Report 4006, INRIA, September 2000.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....in both [17] and eFRCD, the latter considers fault tolerance and real time tasks while the former does not consider either. Very recently, Girault et al. proposed a real time scheduling algorithm for heterogeneous systems that considers fault tolerance and tasks with precedence constraints [5]. This study is by far the closest to eFRCD that the authors have found in the literature. The main differences between [5] and eFRCD are three fold: a) eFRCD considers heterogeneities in computation, communication and reliability that will be defined shortly, whereas the former only considers ....

....either. Very recently, Girault et al. proposed a real time scheduling algorithm for heterogeneous systems that considers fault tolerance and tasks with precedence constraints [5] This study is by far the closest to eFRCD that the authors have found in the literature. The main differences between [5] and eFRCD are three fold: a) eFRCD considers heterogeneities in computation, communication and reliability that will be defined shortly, whereas the former only considers computational heterogeneity. These hetero geneities. b) The former does not take reliability cost into consideration, ....

[Article contains additional citation context not shown here]

A. Girault, C. Lavarenne, M. Sighireanu and Y. Sorel, "Fault-Tolerant Static Scheduling for Real-Time Distributed Embedded Systems," In Proc. of the 21st International Conference on Distributed Computing Systems(ICDCS), Phoenix, USA, April 2001.

....tasks are independent from one another and (2) they are designed only for homogeneous systems. Although heterogeneous systems are considered in both [28] and eFRCD, the latter considers fault tolerance and real time tasks while the former does not consider either. Very recently, Girault et al. [10,11] proposed a real time scheduling algorithm for heterogeneous systems that considers fault tolerance and tasks with precedence constraints. This study is by far the closest to eFRCD that the authors have found in the literature. The main difference between [10,11] and eFRCD are four fold: a) the ....

....Very recently, Girault et al. 10,11] proposed a real time scheduling algorithm for heterogeneous systems that considers fault tolerance and tasks with precedence constraints. This study is by far the closest to eFRCD that the authors have found in the literature. The main difference between [10,11] and eFRCD are four fold: a) the former does not consider task deadlines explicitly, while eFRCD does; b) eFRCD considers heterogeneities in computation, communication and reliability while the former only considers computational heterogeneity; c) the former does not take reliability cost ....

[Article contains additional citation context not shown here]

A. Girault, C. Lavarenne, M. Sighireanu and Y. Sorel, "Fault-Tolerant Static Scheduling for Real-Time Distributed Embedded Systems," In Proc. of the 21st International Conference on Distributed Computing Systems(ICDCS), Phoenix, USA, April 2001. 27

No context found.

A. Girault, C. Lavarenne, M. Sighireanu, and Y. Sorel. Faulttolerant static scheduling for real-time distributed embedded systems. Research Report 4006, INRIA, September 2000.

....processors observe the activity to detect the failure of the main processor. With point topoint links, the detection of the main processor s failure is similar to a Byzantine agreement problem [21] To deal with point to point links and to avoid heavy agreement algorithms, we have proposed in [14] another solution, based on the active redundancy on both comps and comms. In this solution, each operation is replicated times and each replica sends its results to each replica of each successor operation. The idea is that each operation waits until it receives its first set of inputs ....

....since there is a risk that the sum of timeouts amassed overtakes other timeouts. As already said, the current solution is easier and cheaper to implement for architectures where the communication units are connected to a unique multi point link. With point to point links, the solution presented in [14] should be preferred. It should be noted that this other solution also supports several failure occurrences during the same iteration. If the given target architecture uses a single multi point link (e.g. a bus) then the outputs of all operations will be broadcasted over this multi point link ....

A. Girault, C. Lavarenne, M. Sighireanu, and Y. Sorel. Faulttolerant static scheduling for real-time distributed embedded systems. Research Report 4006, INRIA, September 2000.

....detect the processor failures; there is no need for the processors to propagate the state of the faulty ones; and finally, due to the scheduling strategy used the time needed for handling a failure is minimal. A different version of the method presented here has been published as an abstract in [12] and as a full version in a workshop [13] It is different since it addresses distributed architectures consisting of several nodes connected to a single bus, while here we address more general distributed architectures since they can include point to point communication links (see Section 3.3) ....

....point to point communication links (see Section 3.3) As a result, here the communications can be scheduled in parallel on the communication links, and the fault tolerance is achieved with the software redundancy of both the computation operations and the data communications (see Section 4. 1) In [12, 13] we used the time redundancy of the data communications. Also, we can cope with intermittent processor failures and we do not need to use timeouts to detect failures, which was not the case in [12, 13] In conclusion, the method presented here is complementary and more general than the one ....

[Article contains additional citation context not shown here]

A. Girault, C. Lavarenne, M. Sighireanu, and Y. Sorel. Faulttolerant static scheduling for real-time distributed embedded systems. In 21st International Conference on Distributed Computing Systems, ICDCS'01, pages 695--698, Phoenix, USA, April 2001. IEEE. Extended abstract.

....to be taken into consideration. Since failures cannot, by their nature, be predicted, the very basic assumption for off line scheduling, the determinism, is demolished and it seems that this technique should leave space to the other [4] But, as some other studies have shown this is not the case [2, 5, 1]: a certain degree of nondeterminism can be permitted at the scheduling time within the system. However these studies have focused on processor failures, assuming restricted architecture graphs with reliable channels [2, 5] or independent tasks [1] On the other hand, the studies which focus on ....

....other [4] But, as some other studies have shown this is not the case [2, 5, 1] a certain degree of nondeterminism can be permitted at the scheduling time within the system. However these studies have focused on processor failures, assuming restricted architecture graphs with reliable channels [2, 5] or independent tasks [1] On the other hand, the studies which focus on channel failures tend to consider only on line scheduling since they are naturally connected to communication protocols [8, 9, 11] We investigate here the problem of off line fault tolerant scheduling, where both processors ....

A. Girault, C. Lavarenne, M. Sighireanu, and Y. Sorel. Faulttolerant static scheduling for real-time distributed embedded systems. Research Report 4006, INRIA, September 2000. Submitted to IEEE Trans. on Parallel and Distributed Systems.

....while the backup processors observe the bus to detect the failure of the main processor. With point to point links, the detection of the main processor s failure is similar to a Byzantine agreement [5] To deal with point to point links and to avoid heavy agreement algorithms, we have proposed in [2] another solution, based on the active redundancy on both comps and comms. In this solution there is no main replica to choose and no timeout to compute, but on the other hand, there is more communication overhead. 2. How are computed the timeouts associated to the communications Each timeout is ....

....since there is a risk that the sum of timeouts amassed overtakes other timeouts. As already said, the current solution is easier and cheaper to implement for architectures where the communication units are connected to a unique multi point link. With point to point links, the solution presented in [2] should be preferred. It also supports several failures during the same iteration. 4 Conclusion We have presented a scheduling heuristic for obtaining a static distributed fault tolerant schedule, starting from an algorithm specification, an architecture specification, some real time ....

A. Girault, C. Lavarenne, M. Sighireanu, and Y. Sorel. Faulttolerant static scheduling for real-time distributed embedded systems. Research Report 4006, INRIA, September 2000.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC