Demers A et al. Epidemic algorithms for replicated data management. Proceedings of the 6th ACM Symposium on Principles of Distributed Computing (PODC '87), Vancouver, British Columbia, Canada, August 1997. Association for Computing Machinery: New York, NY, 1997; 1--12.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....that researchers are starting to look for solutions that allow very different strategies to co exist in a single system. Most research has concentrated on supporting a single family of strategies. For example, the TACT toolkit [41] provides support for replication based on anti entropy schemes [10] for a range of consistency models. In a similar fashion, caching algorithms exist that base their decisions on temporal correlations between requests [16, 36] but otherwise essentially follow the same protocol. Closer to our approach are systems that have protocols that adapt the way updates are ....

Alan Demers, Dan Greene, Carl Hauser, Wes Irish, John Larson, Scott Shenker, Howard Sturgis, Dan Swinehart, and Doug Terry, Epidemic Algorithms for Replicated Data Management, Sixth Symp. on Principles of Distributed Computing (Vancouver), ACM, August 1987, pp. 1--12.

....if one wants to build robust, reliable, secure distributed systems that can be the basis for the mission critical enterprise systems. But the picture is not entirely bleak. After presenting these arguments, we shift attention to a new class of protocols based on an idea from Clearinghouse [2], the database replication technology developed at Xerox Parc in the 1980 s, and later used by Golding for the refDBMS system [5] and from NNTP, the gossip based algorithm used to propagate Usenet news in the Internet. These turn out to be scalable under the same style of analysis that predicts ....

Demers, A. et. al. Epidemic Algorithms for Replicated Data Management. Proceedings of the 6th Symposium on Principles of Distributed Computing (PODC), Vancouver, Aug. 1987, 1-12.

....if one wants to build robust, reliable, secure distributed systems that can be the basis for the mission critical enterprise systems. But the picture is not entirely bleak. After presenting these arguments, we shift attention to a new class of protocols based on an idea from Clearinghouse [2], the database replication technology developed at Xerox Parc in the 1980 s, and later used by Golding for the refDBMS system [5] and from NNTP, the gossip based algorithm used to propagate Usenet news in the Internet. These turn out to be scalable under the same style of analysis that predicts ....

Demers, A. et. al. Epidemic Algorithms for Replicated Data Management. Proceedings of the 6th Symposium on Principles of Distributed Computing (PODC), Vancouver, Aug. 1987, 1-12.

....not for larger networks. We see management of the multicast dissemination routes for pbcast as a topic for which additional study will be required. 2 Phase Anti Entropy Protocol The important properties of pbcast stem from its gossip based anti entropy protocol. The term anti entropy is from [DGHI87] and refers to protocols that detect and correct inconsistencies in a system by continuous gossiping. Our anti entropy protocol progresses through rounds in which members randomly choose other members, send a summary of their message histories to the selected process, and then solicit copies of ....

A. Demers et. al. Epidemic Algorithms for Replicated Data Management. Proceedings of the 6 Symposium on Principles of Distributed Computing. (Vancouver, CA; Aug. 1987) 112. Also Operating Systems Review 22:1 (Jan. 1988), 832.

....is no differentiation between files. Weaker consistency models have also been adopted. For example, in Ficus, each workstation effectively acts as a server and potential replica host for the files that a client wants to access [17] Updates are propagated in a lazy fashion using anti entropy [7], but because they can be initiated concurrently at different sites, there is a need for reconciliation when write write conflicts occur. Likewise, Coda supports disconnected operation allowing a client to continue updating its local copy of a file, without any guarantees that another client may ....

A. Demers, D. Greene, C. Hauser, W. Irish, J. Larson, S. Shenker, H. Sturgis, D. Swinehart, and D. Terry. "Epidemic Algorithms for Replicated Data Management. " In Proc. Sixth Symp. on Principles of Distributed Computing, pp. 1--12, Vancouver, Aug. 1987. ACM.

....should subscribe to that subject. The underlying communication system ensures that published messages are delivered to their subscribers. The technique for this matching is either based on network level multicasting (as in TIBCO Rendezvous [33] application level multicasting [3] flooding [8, 18], or by means of a network of brokers (as in IBM MQSeries [15] Obviously, each of these approaches has its own scalability problems. A second approach to support anonymous communication and coordination is to make use shared dataspaces that are based on generative communication [13] such as ....

A. Demers et al. "Epidemic Algorithms for Replicated Data Management." In Proc. Sixth Symp. on Principles of Distributed Computing, pp. 1--12, Vancouver, Aug.

....subject should subscribe to that subject. The underlying communication system ensures that published messages are delivered to their subscribers. The technique for this matching is either based on network level multicasting (as in TIBCO Rendezvous [33] application level multicasting [3] ooding [8, 18], or by means of a network of brokers (as in IBM MQSeries [15] Obviously, each of these approaches has its own scalability problems. A second approach to support anonymous communication and coordination is to make use shared dataspaces that are based on generative communication [13] such as ....

A. Demers et al. \Epidemic Algorithms for Replicated Data Management." In Proc. Sixth Symp. on Principles of Distributed Computing, pp. 1-12, Vancouver, Aug.

....that researchers are starting to look for solutions that allow very di erent strategies to co exist in a single system. Most research has concentrated on supporting a single family of strategies. For example, the TACT toolkit [40] provides support for replication based on anti entropy schemes [9] for a range of consistency models. In a similar fashion, caching algorithms exist that base their decisions on temporal correlations between requests [15, 35] but otherwise essentially follow the same protocol. Closer to our approach are systems that have protocols that adapt the way updates are ....

Alan Demers, Dan Greene, Carl Hauser, Wes Irish, John Larson, Scott Shenker, Howard Sturgis, Dan Swinehart, and Doug Terry, Epidemic Algorithms for Replicated Data Management, Sixth Symp. on Principles of Distributed Computing (Vancouver), ACM, August 1987, pp. 1-12.

....should subscribe to that subject. The underlying communication system ensures that published messages are delivered to their subscribers. The technique for this matching is either based on network level multicasting (as in TIBCO Rendezvous [33] application level multicasting [3] flooding [8, 18], or by means of a network of brokers (as in IBM MQSeries [15] Obviously, each of these approaches has its own scalability problems. A second approach to support anonymous communication and coordination is to make use shared dataspaces that are based on generative communication [13] such as ....

A. Demers et al. "Epidemic Algorithms for Replicated Data Management." In Proc. Sixth Symp. on Principles of Distributed Computing, pp. 1--12, Vancouver, Aug.

....the corresponding interfaces will not change often. In addition, it is not necessary that the SAs always share the exact same view of available services. The combination of relative stability and weak consistency allows us to use highly scalable techniques such as deployed in epidemic algorithms [3, 14]. Encapsulating Distribution by Remote Objects 5 Locator The Locator locates an object with a specified CORE OID on behalf of a client, and takes care of binding the client to the requested object. Binding is accomplished by returning a serialized proxy that implements the interface of the ....

A. Demers et al. "Epidemic Algorithms for Replicated Data Management." In Proc. Sixth Symp. on Principles of Distributed Computing, pp. 1--12, Vancouver, Aug. 1987. ACM.

No context found.

Demers A et al. Epidemic algorithms for replicated data management. Proceedings of the 6th ACM Symposium on Principles of Distributed Computing (PODC '87), Vancouver, British Columbia, Canada, August 1997. Association for Computing Machinery: New York, NY, 1997; 1--12.

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