Architecture and protocol support for providing consensus as a fault-tolerant virtualised service

Authors:
Qurat-ul-Ain I. Tariq
Affiliations:
NED University of Engineering & Technology, Karachi, Pakistan
Venue:
Proceedings of the 8th International Conference on Frontiers of Information Technology
Year:
2010

Citing 12
Cited 0

Consensus in the presence of partial synchrony

Journal of the ACM (JACM)
Congestion avoidance and control

SIGCOMM '88 Symposium proceedings on Communications architectures and protocols
Impossibility of distributed consensus with one faulty process

Journal of the ACM (JACM)
Unreliable failure detectors for reliable distributed systems

Journal of the ACM (JACM)
Implementing Fail-Silent Nodes for Distributed Systems

IEEE Transactions on Computers
Practical Byzantine fault tolerance

OSDI '99 Proceedings of the third symposium on Operating systems design and implementation
Fail-stop processors: an approach to designing fault-tolerant computing systems

ACM Transactions on Computer Systems (TOCS)
Group communication specifications: a comprehensive study

ACM Computing Surveys (CSUR)
Randomized Multivalued Consensus

ISORC '01 Proceedings of the Fourth International Symposium on Object-Oriented Real-Time Distributed Computing
Separating agreement from execution for byzantine fault tolerant services

SOSP '03 Proceedings of the nineteenth ACM symposium on Operating systems principles
Randomized Intrusion-Tolerant Asynchronous Services

DSN '06 Proceedings of the International Conference on Dependable Systems and Networks
A Performance Study on the Signal-On-Fail Approach to Imposing Total Order in the Streets of Byzantium

DSN '06 Proceedings of the International Conference on Dependable Systems and Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper addresses the requirements for exporting consensus as a fault-tolerant, virtualised service that is accessible via the Internet to distributed process groups. It then introduces a new concept of Failure Signaler Oracle and presents a consensus protocol that makes use of that oracle to shield the end users from problems afflicting known consensus protocols. Our protocol allows the performance of the resulting consensus service to be decided solely, as in many other e-Services, by the prevailing communication delays and having to cope with real failures. We show how traditional method of introducing redundancy can be used to implement Failure Signaler. It is suggested that minor changes in the protocol design can improve degree of fault-tolerance of the consensus service.