A. Fekete, F. Kaashoek and N. Lynch "Providing Sequentially-Consistent Shared Objects Using Group and Point-to-point Communication" in Proc. of IEEE International Conference on Distributed Computer Systems, 1995, pp 439--449.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....and it is easier to transform a serial program into a distributed shared memory based program, than to a message passing based program. For these reasons and others, much research has been done on how to define and implement distributed shared memory, both from the theoretical point of view [2, 3, 6, 10, 11, 13, 22, 26, 27, 28] and from the practical point of view [14, 17, 18, 19, 21, 31, 32, 33] However, most of this work ignores the possibility of failures and does not allow processors to join an ongoing computation, in order to speed it up. On the other hand, it is clear that if distributed shared memory is to be ....

....and Welch. For reasons of fault tolerance, the algorithm would have to make sure that each memory location is always being held by more than one mcs. Also, such algorithms would need a more elaborated state transfer phase, similar to the one discussed in Section 4.1. However, as been noted by [22], such a solution for sequential consistency may require that each send event, which is labeled as atomic, will be labeled as causal too. 4.3 Causal Memory The code for implementing causal memory appears in Figure 4. Here, too, we assume that each process has a copy of the entire memory. It is ....

A. Fekete, F. Kaashoek, and N. Lynch. Providing Sequentially Consistent Shared Objects Using Group and Point-To-Point Communication. In Proc. 15th International Conference on Distributed Computing Systems, 1995. To appear.

.... head of pending [p] append ha; pi to queue Output brcv(a) p;q Precondition: queue(next [q] ha; pi Effect: next [q] next [q] 1 Figure 3: TO machine The finite traces of this automaton are exactly the finite prefixes of traces of a totally ordered causal broadcast layer, as defined in [26]. Note that, in any trace of TO machine, there is a natural correspondence between brcv events and the bcast events that cause them. 3.2 The Performance and Fault Tolerance Property TO property Consider a signature TO fsig that is the same as that of TO machine, above, with the addition of the ....

A. Fekete, F. Kaashoek and N. Lynch "Providing Sequentially-Consistent Shared Objects Using Group and Point-to-point Communication" in Proc. of IEEE International Conference on Distributed Computer Systems, 1995, pp 439--449.

....each p there is an integer next [p] giving the index in queue of the next data value to be delivered at p. The formal automaton definition is given in Figure 2. The finite traces of this automaton are exactly the finite prefixes of traces of a totally ordered causal broadcast layer, as defined in [20]. Note that, in any trace of TO machine, there is a natural correspondence between brcv events and the bcast events that cause them. 3.2 The Performance and Fault Tolerance Property TO property Consider a signature TO fsig that is the same as that of TO machine, above, with the addition of the ....

A. Fekete, F. Kaashoek and N. Lynch "Providing Sequentially-Consistent Shared Objects Using Group and Point-to-point Communication" in Proc. of IEEE International Conference on Distributed Computer Systems, 1995, pp 439--449.

.... remove head of pending[p] append ha; pi to queue brcv(a) p;q Precondition: queue(next[q] ha; pi Effect: next[q] next[q] 1 Figure 2: TO machine The finite traces of this automaton are exactly the finite prefixes of traces of a totally ordered causal broadcast service, as defined in [14]. Note that, in any trace of TO machine, there is a natural correspondence between brcv events and the bcast events that cause them. Now we define the performance fault tolerance property TO prop. Its signature TO fsig is the same as the signature of TO machine, with the addition of the ....

A. Fekete, F. Kaashoek and N. Lynch "Providing SequentiallyConsistent Shared Objects Using Group and Point-to-point Communication " in Proc. of IEEE International Conference on Distributed Computer Systems, 1995, pp 439--449.

No context found.

A. Fekete, F. Kaashoek and N. Lynch "Providing Sequentially-Consistent Shared Objects Using Group and Point-to-point Communication" in Proc. of IEEE International Conference on Distributed Computer Systems, 1995, pp 439--449.

....each p there is an integer next [p] giving the index in queue of the next data value to be delivered at p. The formal automaton definition is given in Figure 2. The finite traces of this automaton are exactly the finite prefixes of traces of a totally ordered causal broadcast layer, as defined in [14]. Note that, in any trace of TO machine, there is a natural correspondence between brcv events and the bcast events that cause them. TO machine : Signature: Input: bcast(a)p , a 2 A, p 2 P Output: brcv(a) p;q , a 2 A, p; q 2 P Internal: to order(a; p) a 2 A, p 2 P States: queue, a ....

A. Fekete, F. Kaashoek and N. Lynch "Providing Sequentially-Consistent Shared Objects Using Group and Point-to-point Communication" in Proc. of IEEE International Conference on Distributed Computer Systems, 1995, pp 439--449.

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