L. Rodrigues, Ellen Siegel, and P. Verssimo. A Replication-Transparent Remote Invocation Protocol. In Proceedings of the 13th Symposium on Reliable Distributed Systems, Dana Point, California, October 1994. IEEE.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....and between a client and the server, can be weakened. In addition, techniques supporting reliable large scale remote server access have been deployed, allowing services to be easily replicated and invoked transparently, without necessarily implying a degradation of strong failure semantics [Rodrigues et al. 1994]. Both approaches are easily implemented using group based open loop mechanisms. Another style of interaction is multipeer, conveying the notion of spontaneous, symmetric interchange of information, amongst a collection of peer entities. This paradigm appeared as early as in [Powell et al. 1988] ....

L. Rodrigues, E. Siegel and P. Verissimo, "A ReplicationTransparent Remote Invocation Protocol", in Proc. 13th Int. Conf. on FaultTolerant Computing Systems (FTCS-13), (Dana Point, California, USA), IEEE CS, 1994.

....In our protocols, replication concerns are kept local: just in the server group in Protocol 3, and among servers and service clients in Protocol 2. An earlier attempt to propose a Client Access protocol that is independent from the actual Replication mechanism has been made in the GRIP protocol [25]. GRIP has focused on the specific case of the open model, where clients accommodate replication related stubs. However, the functionality of the Client Access protocol is not clearly separated from that of the Replication protocol, especially in the case where at most once execution ....

L. Rodrigues, E. Siegel, and P. Verissimo. A replication-- transparent remote invocation protocol. In Proc. of the 13th IEEE Symposium on Reliable Distributed Systems. Dana-Point, October 1994.

....system. In our protocols, replication concerns are kept local: just in the server group in Protocol 3, and to the service clients in Protocol 2. An earlier attempt to propose client access protocols that are independent from the actual replication mechanism has been made in the GRIP protocol [18]. GRIP focused on the specific case of the open model, where clients accommodate replication related stubs; a protocol similar to 2 has been proposed. However, the functionality of the client access protocol is not clearly separated from that of the replication protocol, especially in the case ....

L. Rodrigues, E. Siegel, and P. Verissimo. A replication-- transparent remote invocation protocol. In Proc. of the 13th IEEE Symposium on Reliable Distributed Systems. DanaPoint, October 1994.

....from a client to a server process is structured to give synchronization semantics at the client similar to normal procedure call. Numerous examples of different RPC services and implementations exist, including Firefly RPC [SB90] Alphorn [AGH 91] lightweight RPC [BALL90] Peregrine [JZ93] RSV94] and SUPRA RPC [Sto94] Among the commercial RPC packages released have been Courier from Xerox [Xer81] Sun RPC [Sun88] Netwise RPC from Novell Netware, and NCA from Apollo [Apo89] TA90] gives a survey of work in this area. On the surface, the semantics of RPC seem very simple, yet the ....

L. Rodrigues, E. Siegel, and P. Verissimo. A replication-transparent remote invocation protocol. In Proceedings of the 13th Symposium on Reliable Distributed Systems, pages 160--169, Dana Point, CA, Oct 1994.

....focused on server fault tolerance only: they do not not even mention client fault tolerance. This approach is like considering only the case of a single invoker invoking multiple invokees (cf Section 2. 4) Systems like AMOEBA [Kaashoek 91] ISIS [Birman 93] HORUS [van Renesse 92] and GRIP [Rodrigues 94] provide support for process groups. A replicated server is implemented using a group of server processes. Depending on the system, server replication is more or less transparent to client processes. AMOEBA, ISIS and HORUS provide primitives for group management and group communication. Using ....

L. Rodrigues, E. Siegel & P. Ver ' issimo. A Replication-Transparent Remote Invocation Protocol. In Proceedings of the 13th Symposium On Reliable Distributed Systems. IEEE, October 1994.

....its hierarchical nature to improve the performance or reliable communication protocols. There are a few published protocols exploiting the innovative attributes of NAVTECH , namely topology and group model. For example, a lightweight group service[36] and a reliable remote group access protocol[37], both taking advantage from the sender member and site participant duality of the group model, and benefit from the topology to have credible efficiency. Another protocol is a clock synchronization protocol based on an hierarchical composition of a protocol tailored to broadcast LANs with a ....

L. Rodrigues, Ellen Siegel, and P. Verssimo. A Replication-Transparent Remote Invocation Protocol. In Proceedings of the 13th Symposium on Reliable Distributed Systems, Dana Point, California, October 1994. IEEE.

....and between a client and the server, can be weakened. In addition, techniques supporting reliable large scale remote server access have been deployed, allowing services to be easily replicated and invoked transparently, without necessarily implying a degradation of strong failure semantics [82]. Both approaches are easily implemented using group based open loop mechanisms. 1.4.2 Multipeer Another style of interaction is multipeer, conveying the notion of spontaneous, symmetric interchange of information, amongst a collection of peer entities. This paradigm appeared as early as in [75] ....

L. Rodrigues, E. Siegel, and P. Verssimo. Replication-transparent remote invocation protocol. In Proceedings of the 13th IEEE Symposium on Reliable Distributed Systems, Dana Point (CA), USA, 1994.

....applications over large scale networks. It offers a range of communication services such as reliable communication based on flexible failure suspectors and partition resilient group membership management complemented with activity support services such as replication management protocols [22, 21]. These services are offered through a flexible interface, allowing client server remote operations, message dissemination, and peer to peer conversations. As part of the validation path for our design options, we are developing a groupware platform. The services of NAVTECH we are concerned with ....

L. Rodrigues, E. Siegel, and P. Verssimo. A Replication-Transparent Remote Invocation Protocol. In Proceedings of the 13th Symposium on Reliable Distributed Systems, October 1994.

....match user requirements. Examples of such primitives are the virtually synchronous communication defined in [13] that only orders bags of messages in respect to group changes) or the global flush protocol of [1] that orders messages in respect to special flush messages. In another report [12], we proposed a primitive that provides the users with explicit control over (message) piggybacking policies. Transparent causal messages One of the criticisms made of causal multicast communication systems [5] concerns the mandatory requirement for reliability. Once a message introduces a A ....

....the request and issues a totally ordered multicast in the group of replicas. In a large scale system, and because communication delays are not transitive, this multicast can be received at R1 before the first write.1 R1 message. With an appropriate retry detection mechanism (for instance, see [12]) it is possible to make immediate progress. Clearly, is not desirable that the multicast depends on the first write request, as this would block replica R1 until the request arrives. However, as in the previous example, it is useful that the precedence relation multicast read.2 R1 is preserved ....

[Article contains additional citation context not shown here]

L. Rodrigues, Ellen Siegel, and P. Verssimo. A Replication-Transparent RemoteInvocation Protocol. In Proceedings of the 13th Symposium on Reliable Distributed Systems, Dana Point, California, October 1994. (To Appear).

....only via protocols tailored to its specific replication protocol [8, 19] causing an undesirable violation of both transparency and modularity. In order to maintain the transparency of an object replication scheme, we have designed a fault tolerant Generic Remote Invocation Protocol (GRIP) [24] which is independent of the service replication protocol and which places no constraints on the replica consistency model. Clients use a lightweight remote access protocol which allows them to remain independent of the details of the replication protocol and of the resulting inter replica ....

L. Rodrigues, E. Siegel, and P. Ver'issimo. A replication-transparent remote invocation protocol. In Proceedings of the 13th Symposium on Reliable Distributed Systems, Dana Point (CA), U.S.A., Oct. 1994.

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