L. Lamport. The part-time parliament. ACM Trans. Comput. Syst., 16(2):133--169, 1998. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
The Need for Realistic Failure Models in Protocol Design - Keidar, Marzullo (2002) (1 citation) (Correct)
....applied to the design of algorithms for real critical systems, e.g. air trac control [6] other highly available services like le servers [15] and so on. It is such a common assumption that most fault tolerant algorithms found in the literature today adopt it without any justi cation (e.g. [14, 19]) It is a common assumption because the t out of n model gives one a simple abstraction for reasoning about failure prone environments and system reliability. With this assumption it is fairly easy to design and verify protocols and also to express lower and upper bounds. Unfortunately, when ....
L. Lamport. The part-time parliament. ACM Trans. Comput. Syst., 16(2):133-169, May 1998.
Challenges in Evaluating Distributed Algorithms - Keidar (2002) (1 citation) (Correct)
....re ned metric can then be used to revisit known lower and upper bound results. e.g. 14] presents a tight lower bound of two communication steps for failure free executions of consensus in practical models. Under the more re ned metric, the lower bound is 2 1 , whereas known algorithms (e.g. [16, 7]) achieve running times of 2 3 . 4 Conclusions Gathering data about Internet characteristics in general, and the behavior of distributed algorithms over the Internet in particular, is extremely important. Such data can be at the basis of more realistic theoretical complexity metrics, and ....
L. Lamport. The part-time parliament. ACM Trans. Comput. Syst., 16(2):133-169, May 1998.
WS-Membership - Failure Management in a Web-Services World - Vogels, Re (2003) (3 citations) (Correct)
....[5] Consensus is used when a set of processes need to agree upon the outcome of an operation. Many consensus protocols require knowledge about which processes are involved in the execution of the protocol to establish a notion of majority, quorums, etc. Even though some protocols such as Paxos [11,12] do not use failure detectors in their specification, the implementation of these protocols is greatly simplified by their use. This paper is organized as follows: section 2 provides a brief overview of the research context in which these services are being developed. In section 3 the membership ....
Lamport, L., "The Part-Time Parliament", ACM Transactions on Computer Systems 16, 2 (May 1998), 133- 169.
SCRIBE: A large-scale and decentralized.. - Castro, Druschel, .. (2002) (117 citations) (Correct)
....must be bu#ered for an amount of time that exceeds the maximal time to repair the multicast tree after a TCP connection breaks. To tolerate root failures, the root needs to be replicated. For example, one could choose a set of replicas in the leaf set of the root and use an algorithm like Paxos [21] to ensure strong consistency. IV. Experimental evaluation This section presents results of simulation experiments to evaluate the performance of a prototype Scribe implementation. These experiments compare the performance of Scribe and IP multicast along three metrics: the delay to deliver ....
L. Lamport, "The Part-Time Parliament," Report Research Report 49, Digital Equipment Corporation Systems Research Center, Palo Alto, CA, Sept. 1989.
Broadcasting Messages in Fault-Tolerant Distributed.. - Charron-Bost, Defago, .. (2002) (Correct)
.... existence of failure detectors of a type that allows us to solve consensus, e.g. the failure detector #S [5] We also assume a majority of correct processes (needed to solve consensus with #S) The rotating coordinator paradigm is well known in the context of fault tolerant computing, e.g. [6, 11, 8, 5]. An rk view is a sequence of processes. The coordinator is the first process in the rk view. a) V R BROADCAST: The implementation of V R BROADCAST is similar to the one of the classical VSC primitive: this can be done by reduction to a consensus problem [12, 9] Here the difference is ....
L. Lamport. The part-time parliament. TR 49, Digital SRC, September 1989.
Fundamental Study Revisiting the PAXOS algorithm - De Prisco, Lampson, Lynch (2000) (2 citations) (Correct)
....the above practical setting. Moreover, the basic safety properties must not be a#ected by the occurrence of failures. Also, the performance of the algorithm must be good when there are no failures, while when failures occur, it is reasonable to not expect e#ciency. Lamport s PAXOS algorithm [19] meets these requirements. The model considered is a partially synchronous distributed system where each process has a direct communication channel with each other process. The failures allowed are timing failures, loss, duplication and reordering of messages, and process stopping failures. ....
....is also achieved and the performance of the algorithm is good. Hence PAXOS has good fault tolerance properties and when the system is stable it combines those fault tolerance properties with the performance of an e#cient algorithm, so that it can be useful in practice. In the original paper [19], the PAXOS algorithm is described as the result of discoveries of archaeological studies of an ancient Greek civilization. That paper contains also a proof of correctness and a discussion of the performance analysis. The style used for the description of the algorithm often diverts the reader s ....
[Article contains additional citation context not shown here]
L. Lamport, The part-time parliament, ACM Trans. Comput. Systems 16 (2) (1998.
Eventually Consistent Failure Detectors - Larrea, Fernández.. (2000) (1 citation) (Correct)
....have been other approaches to solve the Consensus problem in non synchronous systems. In [5] partially synchronous models are assumed and Consensus algorithms for these models have been proposed. Again, these algorithms use the rotating coordinator paradigm and can present the above problem. In [11], another system model is used, in which periods of synchrony and asynchrony alternate. The Paxos Consensus algorithm presented in that work proceeds in asynchronous rounds, 1 In fact, Chandra et al. showed in [3] that a class of failure detectors they denote by 3W is the weakest class required ....
....ones is the way the coordinator is selected. We do not use the rotating coordinator paradigm as they do, but the leader election capability of the new class of failure detectors 3C. In this, it has some similarities with the leader election part of the Paxos Consensus algorithm proposed by Lamport [11]. However, the rest of our algorithm is much closer to the 3S Consensus algorithm presented in [4] Our algorithm also di ers from the Paxos algorithm in the system model assumed (existence of failure detectors versus periods of synchrony) The rest of the paper is organized as follows. In ....
L. Lamport. The part-time parliament. ACM Trans. on Computer Systems, 16(2):133-169, May 1998.
Totally Ordered Broadcast In The Face Of Network Partitions -.. - Keidar, Dolev (1999) (6 citations) (Correct)
....When processes reconnect, the GCS does not recover the states of reconnected processes. This is where the COReL algorithm comes in: COReL recovers lost messages and extends the order achieved by the GCS to a global total order. The majority based Consensus algorithms of Dwork et al. 1988; Lamport, 1989; De Prisco et al. 1997; Chandra and Toueg, 1996; Dolev et al. 1996 are guaranteed to terminate under conditions similar to those of COReL, i.e. at periods at which the network is stable and message delivery is timely, or when failure detectors are eventually accurate. Atomic Broadcast is ....
....is stable and message delivery is timely, or when failure detectors are eventually accurate. Atomic Broadcast is equivalent to Consensus (as proven in Chandra and Toueg, 1996) Atomic Broadcast may be solved by running a sequence of Con sensus decisions (as done, e.g. in Chandra and Toueg, 1996; Lamport, 1989; De Prisco et al. 1997) The main advantage of using COReL over running a sequence of Consensus algorithms is that COReL essentially pipelines the sequence of Consensus decisions. While Consensus algorithms involve special rounds of communication dedicated to exchanging voting messages of the ....
[Article contains additional citation context not shown here]
Lamport, L. (1989). The part-time parliament. TR 49, Systems Research Center, DEC, Palo Alto.
Replicated File Management in Large-Scale Distributed Systems - Babaoglu, Bartoli, Dini (1995) (3 citations) (Correct)
....having to do with lock management, e.g. if the LM left UG. For the sake of clarity, we have presented lock management and writing in isolation. UBLCS 94 16 6 The Replica Management System perform the update or to leave the UG. That is, a participant can never respond NO to a write request [16, 17] as it could in an atomic commitment algorithm [13] In other words, our write protocol does not require an atomic commitment. The reason why we can make this assumption is the observation that a server unable to write to its disks must have serious problems and will eventually be declared ....
L. Lamport. The part-time parliament. Technical Report 49, DEC SRC, Palo Alto, 1989.
Cheap Paxos - Lamport, Massa (2004) Self-citation (Lamport) (Correct)
No context found.
L. Lamport. The part-time parliament. ACM Trans. Comput. Syst., 16(2):133--169, May 1998.
How Fast Can Eventual Synchrony Lead to Consensus? - Dutta, Guerraoui, Lamport (2005) Self-citation (Lamport) (Correct)
No context found.
L. Lamport. The part-time parliament. ACM Trans. Comput. Syst., 16(2):133--169, May 1998.
Quorum Placement in Networks: Minimizing Network.. - Golovin, Gupta, Maggs.. (2006) (Correct)
No context found.
L. Lamport. The part-time parliament. ACM Trans. Comput. Syst., 16(2):133--169, 1998.
FUSE: Lightweight Guaranteed Distributed Failure Notification - John Dunagan Nicholas (2004) (1 citation) (Correct)
No context found.
L. Lamport. The Part-Time Parliament. ACM TOCS, 16:133--169, 1998.
Separating Agreement from Execution for Byzantine.. - Yin, Martin.. (2003) (3 citations) (Correct)
No context found.
L. Lamport. Part time parliament. ACM Trans. on Computer Systems, 16(2), May 1998.
Implementing Trustworthy Services Using Replicated State.. - Schneider, Zhou (2005) (Correct)
No context found.
L. Lamport, "The Part-Time Parliament," ACM Trans. Computer Systems, vol. 16, no. 2, 1998, pp. 133--169.
Separating Agreement from Execution for Byzantine.. - Yin, Martin.. (2003) (3 citations) (Correct)
No context found.
L. Lamport. Part time parliament. ACM Trans. on Computer Systems, 16(2), May 1998.
Using N-Trees for Scalable Event Ordering - Games (Correct)
No context found.
L. Lamport. The part-time parliament. ACM Trans. on Comp. Syst., 16(2):133--169, May 1998.
FUSE: Lightweight Guaranteed Distributed Failure.. - Dunagan, Harvey.. (2004) (1 citation) (Correct)
No context found.
L. Lamport. The Part-Time Parliament. ACM TOCS, 16:133--169, 1998.
Performance Comparison of a Rotating Coordinator.. - Urban.. (2004) (Correct)
No context found.
L. Lamport, "The part-time parliament," ACM Trans. on Computer Systems, vol. 16, no. 2, pp. 133--169, 1998.
Separating Agreement from Execution for Byzantine.. - Yin, Martin.. (2003) (3 citations) (Correct)
No context found.
L. Lamport. Part time parliament. ACM Trans. on Computer Systems, 16(2), May 1998.
FUSE: Lightweight Guaranteed Distributed Failure Notification - John Dunagan Nicholas (2004) (1 citation) (Correct)
No context found.
L. Lamport. The Part-Time Parliament. TOCS, 16:133--169, 1998.
Separating Agreement from Execution for - Zantine Fault Tolerant (Correct)
No context found.
L. Lamport. Part time parliament. ACM Trans. on Computer Systems, 16(2), May 1998.
Separating Agreement from Execution for Byzantine.. - Yin, Martin.. (2003) (3 citations) (Correct)
No context found.
L. Lamport. Part time parliament. ACM Trans. on Computer Systems, 16(2), May 1998.
Practical Impact of Group Communication Theory - Schiper (2003) (1 citation) (Correct)
No context found.
32 L. Lamport. The Part-Time Parliament. Technical Report 49, Digital SRC, September 1989.
ACM SIGACT News Distributed Computing Column 3 - Sergio Rajsbaum May (Correct)
No context found.
L. Lamport. The part-time parliament. ACM Trans. Comput. Syst., 16(2):133-169, May 1998.
ACM SIGACT News Distributed Computing Column 3 - Sergio Rajsbaum May (Correct)
No context found.
L. Lamport. The part-time parliament. ACM Trans. Comput. Syst., 16(2):133-169, May 1998.
Challenges in Evaluating Distributed Algorithms - Keidar (2003) (1 citation) (Correct)
No context found.
16 L. Lamport. The part-time parliament. ACM Trans. Comput. Syst., 16(2):133-169, May 1998.
Open Questions on Consensus Performance - Keidar, Rajsbaum (2003) (Correct)
No context found.
20 L. Lamport. The part-time parliament. ACM Trans. Comp. Sys., 16(2):133{ 169, May 1998.
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