X. Wang, H. Zhao, and J. Zhu, "GRPC: A Communication Cooperation Mechanism in Distributed Systems", ACM Operating System Review, 27(3), 1993.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....the mechanisms for event subscription, and their overall run time architecture (see [13] for a detailed characterization of these aspects) In general, the products and approaches we mention in this section do not support the mobility of the software components exchanging events. Multicast RPC [3, 18, 19] (also known as group RPC) allows a client to invoke a service on a group of servers which exports the same interface. Servers register to a class of messages (service requests) by joining a group and by exporting the common interface defined for the group. This is quite different from the ....

X. Wang, H. Zhao, and J. Zhu, "GRPC: A Communication Cooperation Mechanism in Distributed Systems", ACM Operating System Review, 27(3), 1993.

....communication. Group RPC has numerous applications. For example, it can be used to implement replicated servers to increase availability of the service in the event of failures, to implement parallel computation, or to improve response time. Examples of group or multicast RPC include [7, 8, 23, 25]. For brevity, in this paper we consider only one to many group RPC, in which one client uses RPC to invoke a procedure implemented by a server group. The semantics of group RPC are identical to ordinary RPC when considering the call, orphan handling, communication, and termination semantics ....

....Acceptance with limit equal to the size of the server group, Collation with cumulative function consisting of identity and comparison to detect inconsistencies at the server processes, Reliable Communication, Unique Execution, and Total Order. As an example of a very simple group RPC, lookup RPC [23] corresponds to RPC Main, Synchronous Call, Acceptance with limit one, and Collation with identity. Also, note that the dependency graph given here does not map directly to the dependency graph of the properties given in Figure 2. One difference is that the properties in Figure 2 include those ....

X. Wang, H. Zhao, and J. Zhu. GRPC: A communication cooperation mechanism in distributed systems. Operating Systems Review, 27(3):75--86, Jul 1993.

....Group RPC has numerous applications. For example, it can be used to implement replicated servers to increase availability of the service in the event of failures, to implement parallel computation, or to improve response time. Examples of group or multicast RPC include [Coo85, YJT88, CGR88, WZZ93, Che86, SS90, Mar86, Coo90] For brevity, in this paper we consider only one to many group RPC, in which one client uses RPC to invoke a procedure implemented by a server group. The semantics of group RPC are identical to ordinary RPC when considering the call, orphan handling, communication, and ....

Wang Xingwei, Zhao Hong, and Zhu Jiakeng. GRPC: A communication cooperation mechanism in distributed systems. Operating Systems Review, 27(3):75--86, Jul 1993.

....servers execute all calls. At most once execution is provided, as well as FIFO ordering. All servers responses are identical, but only one response is returned to the client process. Another Sun RPC based group service offers three variations of semantics to support common classes of applications [WZZ93] A lookup style of GRPC dispatches an RPC call to a group of servers and if any of them respond, the call completes. No ordering or execution guarantees are provided, so this is ideal for servers that are stateless and just respond to lookup requests from clients. The functional convergence GRPC ....

X. Wang, H. Zhao, and J. Zhu. GRPC: A communication cooperation mechanism in distributed systems. ACM Op. Syst. Review, 27(3):75--86, July 1993.

....Group RPC has numerous applications. For example, it can be used to implement replicated servers to increase availability of the service in the event of failures, to implement parallel computation, or to improve response time. Examples of group or multicast RPC include [CGR88, Coo85, WZZ93, YJT88] Here, we consider only one to many group RPC, in which one client uses RPC to invoke a procedure implemented by a server group. Other variants not discussed here are many to one RPC, where a replicated client invokes an RPC on a non replicated server, and many to many RPC, where a ....

....invocation terminates normally, the remote procedure has been executed one or more times on at least one server site, and if it terminates abnormally, no conclusion is possible. This semantics is useful for getting one read only response quickly and is used, for example, in the lookup RPC of GRPC [WZZ93] At least once on all semantics is defined similarly, except that the invocation must have been executed on all server sites. Note that the number of sites where the call must succeed can be anything between one and all . Similarly, we can define exactly once on : and at most once on : ....

[Article contains additional citation context not shown here]

X. Wang, H. Zhao, and J. Zhu. GRPC: A communication cooperation mechanism in distributed systems. Operating Systems Review, 27(3):75-- 86, Jul 1993.

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