Epidemic algorithms for replicated database maintenance

@inproceedings{Demers1987EpidemicAF,
  title={Epidemic algorithms for replicated database maintenance},
  author={Alan J. Demers and Daniel H. Greene and Carl H. Hauser and Wes Irish and John Larson and Scott Shenker and Howard E. Sturgis and Daniel C. Swinehart and Douglas B. Terry},
  booktitle={PODC '87},
  year={1987}
}
Whru a dilt~lhSC is replicated at, many sites2 maintaining mutual consistrnry among t,he sites iu the fac:e of updat,es is a signitirant problem. This paper descrikrs several randomized algorit,hms for dist,rihut.ing updates and driving t,he replicas toward consist,c>nc,y. The algorit Inns are very simple and require few guarant,ees from the underlying conllllunicat.ioll system, yc+ they rnsutc t.hat. the off(~c~t, of (‘very update is evcnt,uwlly rf+irt-ted in a11 rq1ica.s. The cost, and… Expand
Epidemic Algorithms for Replicated Databases
TLDR
A variant of the pessimistic algorithm is developed which is optimistic in that transactions commit as soon as they terminate locally and inconsistencies are detected asynchronously as the effects of committed transactions propagate through the system. Expand
Epidemic algorithms in replicated databases (extended abstract)
TLDR
A variant of the pessimistic algorithm in which locks are released as soon as transactions finish their execution locally is developed, however, this optimistic releasing of locks introduces the possibility of cascading aborts while ensuring serializable executions. Expand
Precise Analysis of Epidemic Algorithms
Epidemic algorithms are distributed algorithms in which the agents in thenetwork involve peers similarly to the spread of epidemics. In this work, we focus on randomized rumor spreading -- a class ofExpand
Performance stabilization of a token based epidemic diffusion
TLDR
Stochastic diffusion proves to scale well with system size, while ensuring a high degree of predictability, and is observed that epidemic diffusion tends to insist on domain boundaries. Expand
Replicated document management in a group communication system
TLDR
The design and implementation of a repllwted database that forms the basis for the N&s*gray, communication system, which supports groups of people working on shared sets of donunents and provides end-users the ability to design and create document databases for specfflc applications. Expand
Decentralized Schemes for Size Estimation in Large and Dynamic Groups
TLDR
This paper presents two new decentralized algorithms for estimation in dynamic groups called group size estimation, for estimating the number of non-faulty processes present in the global group at any given point of time. Expand
On Propagating Updates in a Byzantine Environment
TLDR
This work proposes two epidemic-style diffusion algorithms and two measures that characterize the efficiency of diffusion algorithms in general, and characterize both of their algorithms according to these measures, and proves lower bounds with regards toThese measures that show that the authors' algorithms are close to optimal. Expand
Epidemic-style proactive aggregation in large overlay networks
  • M. Jelasity, A. Montresor
  • Computer Science
  • 24th International Conference on Distributed Computing Systems, 2004. Proceedings.
  • 2004
TLDR
The proposed antientropy aggregation protocols compute different aggregates of component properties like extremal values, average and counting, which make them suitable for implementing e.g. collective decision making or automatic system maintenance based on global information in a fully distributed fashion. Expand
On diffusing updates in a Byzantine environment
TLDR
The first analysis of epidemic-style protocols for such environments is provided, fundamentally different from known analyses for the benign case due to the treatment of fully Byzantine failure-which precludes the use of digital signatures for authenticating forwarded updates. Expand
Practical Algorithms for Size Estimation in Large and Dynamic Groups
Large-scale distributed systems may be required to estimate the number of non-faulty processes present in the group at a given point of time. The problem is related to aggregation, and severalExpand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 38 REFERENCES
Weighted voting for replicated data
TLDR
The algorithm guarantees serial consistency, admits temporary copies in a natural way by the introduction of copies with no votes, and has been implemented in the context of an application system called Violet. Expand
Discarding Obsolete Information in a Replicated Database System
TLDR
A protocol is described which allows sites in the system to agree that updates older than a given timestamp are no longer needed and can be discarded, which represents a practical application of the "distributed snapshot" algorithm of Chandy and Lamport. Expand
The clearinghouse: a decentralized agent for locating named objects in a distributed environment
The problem of naming and locating objects in a distributed environment is considered, and the clearinghouse, a decentralized agent for supporting the naming of these "network-visible" objects, isExpand
On spreading a rumor
Suppose that one of n people knows a rumor. At the first stage, he passes the rumor to someone chosen at random; at each stage, each person already informed (“knower”) communicates the rumor to aExpand
Probabilistic solitude verification on a ring
TLDR
This paper is concerned with versions of solitude verification which cannot be solved with certainty, but can be solved probabilisticaUy, with arbitrarily small positive probability of error. Expand
Efficient and reliable broadcast is achievable in an eventually connected network(Extended Abstract)
TLDR
The main idea of the protocol is to perform the broadcast along a dynamic tree, such that the father of each node is the neighbor which has received the maximum number of messages, which guarantees that the tree will eventually reconnect each sub-network to its complement. Expand
Distributed snapshots: determining global states of distributed systems
TLDR
An algorithm by which a process in a distributed system determines a global state of the system during a computation, which helps to solve an important class of problems: stable property detection. Expand
Grapevine: an exercise in distributed computing
TLDR
The set of services provided by Grapevine and how its data and function are divided among computers on the internet are described and described. Expand
Designing a global name service
TLDR
The global name service described here is meant to do this for billions of names distributed throughout the world and addresses the problems of high availability, large size, continuing evolution, fault isolation and lack of global trust. Expand
Another advantage of free choice (Extended Abstract): Completely asynchronous agreement protocols
TLDR
This work exhibits a probabilistic solution for this problem, which guarantees that as long as a majority of the processes continues to operate, a decision will be made (Theorem 1). Expand
...
1
2
3
4
...