Filters








21,372 Hits in 3.6 sec

White-Box Atomic Multicast (Extended Version) [article]

Alexey Gotsman, Anatole Lefort, Gregory Chockler
2019 arXiv   pre-print
In this paper we propose a novel genuine atomic multicast protocol that in the absence of failures takes as low as 3 message delays to deliver a message when no other messages are multicast concurrently  ...  Atomic multicast is a communication primitive that delivers messages to multiple groups of processes according to some total order, with each group receiving the projection of the total order onto messages  ...  We also thank Paulo Coelho and Fernando Pedone for discussions about their FastCast protocol. Alexey Gotsman was supported by an ERC grant RACCOON.  ... 
arXiv:1904.07171v1 fatcat:krws25upzvbgxmgqh53qdzrl4a

Continuous Timestamping for Efficient Replication Management in DHTs [chapter]

Reza Akbarinia, Mounir Tlili, Esther Pacitti, Patrick Valduriez, Alexandre A. B. Lima
2010 Data Management in Grid and Peer-to-Peer Systems  
To perform updates on replicas, we propose a new protocol that stamps update actions with timestamps generated in a distributed fashion.  ...  We propose a new service, called Continuous Timestamp based Replication Management (CTRM), which deals with the efficient storage, retrieval and updating of replicas in DHTs.  ...  of replica holder groups and monotonically increasing and continuous timestamps.  ... 
doi:10.1007/978-3-642-15108-8_4 dblp:conf/globe/AkbariniaTPVL10 fatcat:efgnpk22ybehfcvblb64avrdr4

Hard Real-Time Reliable Multicast in the Dedos System [chapter]

Dick Alstein, Peter Stok
1995 Responsive Computer Systems: Steps Toward Fault-Tolerant Real-Time Systems  
The basic protocol tolerates only crash failures of processes, while the second protocol can also handle timing failures. The protocols can be extended so that multicast group membership is dynamic.  ...  The protocols are shown to be tolerant of memory crashes, and of process or processor failures.  ...  Acknowledgements This paper has benefited from the COllllllents of Dieter Hammer, Henk Schepers, and especially Mathai Joseph, whose remarks have led to a. considerahle reduction of the complexity of the protocols  ... 
doi:10.1007/978-1-4615-2271-3_7 fatcat:gsclcbgfgzgkfigia7er7ykyle

Lightweight causal and atomic group multicast

André Schiper, Kenneth Birman, Pat Stephenson
1991 ACM Transactions on Computer Systems  
that all processes will respect the flush protocol, since the failure of a process pi could prevent it from sending flush messages for some view, even if that view reports pi as still operational.  ...  Delivery atomicity and virtual synchrony when failures occur. We now consider the case where some process fails during an execution.  ... 
doi:10.1145/128738.128742 fatcat:touj52ue6zgqvjizvct6d2was4

The Paxos Register

Harry C. Li, Allen Clement, Amitanand S. Aiyer, Lorenzo Alvisi
2007 Symposium on Reliable Distributed Systems. Proceedings  
values and (ii) tokens that capture how these protocols guarantee agreement despite partial failures.  ...  We use our register to show how Lamport's Classic Paxos and Castro and Liskov's Byzantine Paxos are the same consensus protocol, but for different failure models.  ...  For crash failures, we can implement a simple leader election protocol. For Byzantine failures, we can allot exponentially increasing windows of time to each proposer in a round-robin fashion [5] .  ... 
doi:10.1109/srds.2007.4365689 fatcat:ns2bc5o5kvg7bhszx32qci7tey

The Paxos Register

Harry C. Li, Allen Clement, Amitanand S. Aiyer, Lorenzo Alvisi
2007 2007 26th IEEE International Symposium on Reliable Distributed Systems (SRDS 2007)  
values and (ii) tokens that capture how these protocols guarantee agreement despite partial failures.  ...  We use our register to show how Lamport's Classic Paxos and Castro and Liskov's Byzantine Paxos are the same consensus protocol, but for different failure models.  ...  For crash failures, we can implement a simple leader election protocol. For Byzantine failures, we can allot exponentially increasing windows of time to each proposer in a round-robin fashion [5] .  ... 
doi:10.1109/srds.2007.32 dblp:conf/srds/LiCAA07 fatcat:7y7kz37ylbbk5ew6doozwpqvxq

Analysis and Implementation of the Authentication Protocol 802. 1x

A. I.A.Jabbar, Harith G. A.
2014 International Journal of Computer Applications  
The security of wireless networks against hacker attacks depends on some parameters and special configurations.  ...  It is now the most famous and applicable protocol for authenticating remote users [6] .  ...  Fig 2 : 2 Successful and Failure processes according to the 802.1X protocol International Journal of Computer Applications (0975 -8887) Volume 87 -No.4, February 2014 Figure 3 3 displays some of the  ... 
doi:10.5120/15200-3587 fatcat:kvaenpqwnbctpc2gmaajw2gsdm

Optimal Resilience for Erasure-Coded Byzantine Distributed Storage [chapter]

Christian Cachin, Stefano Tessaro
2005 Lecture Notes in Computer Science  
Additionally, we give the first implementation of non-skipping timestamps which provides optimal resilience and withstands Byzantine clients; it is based on threshold cryptography.  ...  Like some of the previous work, our protocol uses interaction among the servers.  ...  Towards a contradiction, assume there is an adversary and a tag ID such that in some run a write operation has been accepted by some honest server with timestamp ts, but for some ts < ts, no write operation  ... 
doi:10.1007/11561927_39 fatcat:bztnzcnox5cx3ffrsmtnvaew5e

Replication in DHTs Using Dynamic Groups [chapter]

Reza Akbarinia, Mounir Tlili, Esther Pacitti, Patrick Valduriez, Alexandre A. B. Lima
2011 Lecture Notes in Computer Science  
To perform updates on replicas, we propose a new protocol that stamps the updates with timestamps that are generated in a distributed fashion using the dynamic groups.  ...  We propose a new service, called Continuous Timestamp based Replication Management (CTRM), which deals with the efficient storage, retrieval and updating of replicas in DHTs.  ...  In contrast, if there is some gap in the received timestamps, then there should be some missed updates at p.  ... 
doi:10.1007/978-3-642-23074-5_1 fatcat:2ejo52smf5bahfuquozlv65zxe

Minimal Byzantine Storage [chapter]

Jean-Philippe Martin, Lorenzo Alvisi, Michael Dahlin
2002 Lecture Notes in Computer Science  
Our protocol requires 3f + 1 servers to tolerate up to f Byzantine faults-f fewer than the 4f + 1 required by existing protocols for non-self-verifying data.  ...  We show that this protocol is optimal in the number of servers-any protocol that provides safe semantics or stronger requires at least 3f + 1 servers to tolerate f Byzantine faults in an asynchronous system  ...  Hence no later read can return an earlier timestamp. (2) Suppose that at some global time t 1 , some client c reads timestamp ts 1 .  ... 
doi:10.1007/3-540-36108-1_21 fatcat:4pxt7gkztza6dhw7zn443jj3bu

Scalable Transaction Management with Snapshot Isolation for NoSQL Data Storage Systems

Vinit Padhye, Anand Tripathi
2015 IEEE Transactions on Services Computing  
We also describe the cooperative recovery protocol for dealing with transaction failures. A.  ...  A transaction randomly stalls during the commit protocol execution with certain probability called as failure probability. The failure probability is calculated based on the desired failure rate.  ... 
doi:10.1109/tsc.2013.47 fatcat:p5yylbindjdrditzejixvukpk4

Byzantine quorum systems

Dahlia Malkhi, Michael Reiter
1998 Distributed computing  
y despite these failures.  ...  We explore several variations of our quorum systems and extend our constructions to cope with arbitrary client failures.  ...  Our protocols require that different clients choose different timestamps, and thus each client c chooses its timestamps from some set T. that does not intersect T,, for any other client c'.  ... 
doi:10.1007/s004460050050 fatcat:bt46czohfrhm7pbs2e4qkltliu

Renewing Cryptographic Timestamps [chapter]

Sattam S. Al-Riyami, Chris J. Mitchell
2002 IFIP Advances in Information and Communication Technology  
., 1994] for extending the validity of a cryptographic timestamp for a Time Stamping Service contains security shortcomings. A modification is proposed to rectify the identified shortcomings.  ...  Note that we assume R = (MIIX) for some X. h has been subject to a particUlarly severe form of failure, i.e. so that we can find pre-images for which the first part can be freely chosen.  ...  In the attack we suppose that the hashfunction used in the original timestamping protocol has been compromised at some point after the renewal process, i.e. the very situation which renewal is supposed  ... 
doi:10.1007/978-0-387-35612-9_2 fatcat:u7yovpcye5eftozjz4o4423bai

Byzantine quorum systems

Dahlia Malkhi, Michael Reiter
1997 Proceedings of the twenty-ninth annual ACM symposium on Theory of computing - STOC '97  
y despite these failures.  ...  We explore several variations of our quorum systems and extend our constructions to cope with arbitrary client failures.  ...  Our protocols require that different clients choose different timestamps, and thus each client c chooses its timestamps from some set T. that does not intersect T,, for any other client c'.  ... 
doi:10.1145/258533.258650 dblp:conf/stoc/MalkhiR97 fatcat:2s3x4c3tgvambhycmo6wjlqfya

Optimistic distributed simulation based on transitive dependency tracking

Om P. Damani, Yi-Min Wang, Vijay K. Garg
1997 ACM SIGSIM Simulation Digest  
So a single straggler may result in a large number of anti-messages and multiple rollbacks of some LPs. In our protocol, an LP receiving a straggler broadcasts its rollback.  ...  In traditional optimistic distributed simulation protocols, a logical process(LP) receiving a straggler rolls back and sends out anti-messages.  ...  Since the content of volatile memory is lost upon a failure, some of the messages are no longer available for replay after the failure. Thus, some of the process states are lost in the failure.  ... 
doi:10.1145/268823.268906 fatcat:g5jadyqeqneqtn4f4mnjz4g2va
« Previous Showing results 1 — 15 out of 21,372 results