Ken P. Birman. The Proecss Group Appraoch to Reliable Distributed Computing. Communications of the ACM, 36(12):36--53, 1993.  Home/Search   Document Not in Database   Summary   ACM   TOC   Related Articles   Check

....execution. Depending on the specified consistency requirements, a replica may need to contact other replicas before processing a particular request. Replicas exchange updates by propagating writes. This can take the form of gossip messages [22] antientropy sessions [13, 28] group communication [5], broadcast, etc. We chose anti entropy exchange as our write propagation method because of its flexibility in operating under a variety of network scenarios. Each write bears an accept stamp composed of a logical clock time [23] and the identifier of the accepting replica. Replicas ....

Ken P. Birman. The Proecss Group Appraoch to Reliable Distributed Computing. Communications of the ACM, 36(12):36--53, 1993.

....execution. Depending on the specified consistency requirements, a replica may need to contact other replicas before processing a particular request. Replicas exchange updates by propagating writes. This can take the form of gossip messages [17] anti entropy sessions [12, 23] group communication [4], broadcast, etc. We chose antientropy exchange as our write propagation method because of its flexibility in operating under a variety of network scenarios. Each write bears an accept stamp composed of a logical clock time [18] and the identifier of the accepting replica. Replicas ....

Ken P. Birman. The Proecss Group Appraoch to Reliable Distributed Computing. Communications of the ACM, 36(12):36--53, 1993.

....are propagated to the other servers in the WebFS group. This caching model allows for out of order message delivery, but we deemed such semantics to be acceptable for a chat application. If it is determined that such semantics are insufficient, well known distributed techniques [Ladin et al. 1992, Birman 1993] can be used to provide strong ordering of updates. A preliminary prototype of the chat application is up and running. Initial measurements of the system indicate that the system is indeed scalable. One WebFS server propagates approximately 30 updates per second to four clients attempting to ....

....the system [Nichols 1987, Bricker et al. 1991, Douglis Ousterhout 1991, Zhou et al. 1992] Once again, all these systems implement server side solutions for load balancing and require client intervention to spread jobs among cluster machines. Perhaps most closely related to our systems are ISIS [Birman 1993] and so called gossip architectures [Ladin et al. 1992] which provide a substrate for developing distributed applications. ISIS reliable group communication to support the type of applications we envision. Gossip architectures use front ends analogous to Smart Clients to access replicated servers ....

K. P. Birman. "The Proecss Group Appraoch to Reliable Distributed Computing". Communications of the ACM, 36(12):36--53, 1993.

.... Other Web services such as chat rooms and whiteboards have long been structured using multicast; we can simplify these applications by folding the multicast support into the file system (for these applications, we use weak coherence since updates need not arrive in a consistent order at all sites [7]) Both the last writer wins and the multicast update invalidate coherence policies have been implemented in WebFS [54] We also plan to investigate transparent result caching. HTTP servers translate some URLs into requests to dynamically generate a Web page. HTTP servers translate URLs matching a ....

....result caching is used to cache dynamic Web pages and to ensure invalidation of these dynamic pages at the proper time. 6 Related Work We are able to leverage and evaluate a huge research literature in the context of Web services, including: models for building fault tolerant applications [4, 8, 45, 34, 9, 40, 7, 17, 31, 47, 10, 48]; process management across the local area [53, 22, 42, 28, 57, 14] wide area file systems [35, 15, 39, 52, 16, 33, 1] wide area security systems [20, 46, 49, 27, 12, 55, 26] and mechanisms for hiding fault tolerance, load balancing, and dynamic relocation from end users [31, 38, 6, 11, 2] In ....

Ken P. Birman. The Proecss Group Appraoch to Reliable Distributed Computing. Communications of the ACM, 36(12):36--53, 1993.

....chat application. Chat rooms are modeled as files with reads corresponding to receiving conversation updates and writes to sending out a message. On a write, the WebFS updates all its clients; the updates are propagated to other servers in a lazy fashion. tribution techniques [Ladin et al. 1992, Birman 1993] can be used to provide strong ordering of updates. Since the read requests are idempotent, and the write requests are atomic with respect to WebFS, the chat application is completely tolerant to server crashes. This fault transparency provides the illusion of a single, highly available chat ....

....authentication, and linking resource manager object code with the Transaction Processing operating system (TPOS) This lightweight nature enables Smart Clients to be downloaded into existing Web browsers to customize existing Internet services. Also related to our systems are ISIS [Birman 1993] and gossip architectures [Ladin et al. 1992] which provide a substrate for developing distributed applica tions. ISIS provides reliable group communication to support many of the applications we envision. Gossip architectures use front ends analogous to Smart Clients to access replicated ....

K. P. Birman. "The Proecss Group Appraoch to Reliable Distributed Computing". Communications of the ACM, 36(12):36--53, 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