AI helps you reading Science

AI generates interpretation videos

AI extracts and analyses the key points of the paper to generate videos automatically


pub
Go Generating

AI Traceability

AI parses the academic lineage of this thesis


Master Reading Tree
Generate MRT

AI Insight

AI extracts a summary of this paper


Weibo:
We propose the Sigma: a practical, efficient and fault-tolerant protocol for distributed mutual exclusion inside P2P DHT

A practical distributed mutual exclusion protocol in dynamic peer-to-peer systems

IPTPS, (2004): 11-21

Cited: 59|Views73
EI WOS SCOPUS

Abstract

Mutual exclusion is one of the well-studied fundamental primitives in distributed systems. However, the emerging P2P systems bring forward several challenges that can’t be completely solved by previous approaches. In this paper, we propose the Sigma protocol that is implemented inside a dynamic P2P DHT and circumvents those issues. The ba...More

Code:

Data:

0
Introduction
  • One of the fundamental primitives to implement more generic systems and applications on top of P2P DHTs [4][13] is mutual exclusion.
  • For the applications and systems the authors envision to be built and deployed on those P2P DHTs, one can all but rule out the possibility of enforcing concurrency using stable transaction servers, whether they are external or internal to the system.
  • Such primitives must be implemented inside P2P DHT.
  • The protocol is by definition distributed, it must be simple and efficient, and yet robust enough to be of practical use
Highlights
  • One of the fundamental primitives to implement more generic systems and applications on top of P2P DHTs [4][13] is mutual exclusion
  • The Sigma protocol is fully implemented and deployed in a distributed testbed, which can be configured by different network topology models
  • The emerging P2P scenario brings forward several challenges to mutually exclusive access of the resource stored in it, such as the huge variance of network latency, unpredictable number of clients and high dynamism. These issues are partially addressed in previous works but not completely solved
  • We propose the Sigma: a practical, efficient and fault-tolerant protocol for distributed mutual exclusion inside P2P DHT
  • The key points of Sigma protocol are to use logical replicas and quorum consensus to deal with system dynamisms
  • Quasi-consistency and cooperation between clients and replicas circumvent the large variance of network latency and high contention
Results
  • The Sigma protocol is fully implemented and deployed in a distributed testbed, which can be configured by different network topology models.

    The authors assume a pool of infinity clients, and each client will fire request contending for CS according to a Poisson distribution, with as the incoming request rate.
  • After 5 minutes warm-up period the authors test 10 minutes during which throughput, in terms of the number of serviced requests per second, is measured
  • This is repeated for different incoming request rate.
  • (a,b)=(90,110) ms saturated service rate dashed lines correspond to theoretical predictions of saturated throughput, which differ with latency distribution.
  • Data for both the strawman and Sigma are shown.
  • Data for both the strawman and Sigma are shown. m/n is 24/32
Conclusion
  • The emerging P2P scenario brings forward several challenges to mutually exclusive access of the resource stored in it, such as the huge variance of network latency, unpredictable number of clients and high dynamism.
  • These issues are partially addressed in previous works but not completely solved.
  • The key points of Sigma protocol are to use logical replicas and quorum consensus to deal with system dynamisms.
  • Sigma gracefully deals with failure by two techniques: informed backoff and lease, making protocol faulttolerant
Related work
  • From the taxonomy of [10], the Sigma protocol would fall into the “permission-based” category. These protocols assume a closed system, in which clients are also the replicas. The context of this work mandates an open system where number of clients is unpredictable.

    The more relevant work includes the Byzantine protocols [6][9][5] which also operates in an opensystem setting. Obviously, Sigma’s idea of virtual replicas is immediately applicable to these protocols to tailor-fit them in a P2P environment. The objectives, however, differ. Sigma is a light-weight synchronization protocol with O(n) message costs and does not attempt to deal with malicious client. Whereas the Byzantine protocols takes a replicated state machine approach with O(n2) cost and handles malicious client. It is interesting to note that, for a total of 3f+1 replica, when faults exceed f, both protocols will yield unpredictable results.
Reference
  • B. Pawlowski, S. Shepler, et al. The NFS Version 4 Protocol, in Proceedings of the 2nd international system administration and networking conference (SANE2000)
    Google ScholarLocate open access versionFindings
  • D. Liben-Nowell, H. Balakrishnan, and D. Karger, Analysis of the Evolution of Peer-to-Peer Systems, in 21st ACM Symposium on Principles of Distributed Computing (PODC), Monterey, CA, July 2002.
    Google ScholarFindings
  • G. Chokler, D. Malkhi, and M. Reiter, Backoff Protocols for Distributed Mutual Exclusion and Ordering, in Proceedings of 21st International Conference on Distributed Computing Systems (ICDCS), 2001.
    Google ScholarLocate open access versionFindings
  • I. Stoica, et al, Chord: A Scalable Peer-to-peer Lookup Service for Internet Applications, in Proceedings of ACM SIGCOMM 2001, San Deigo, CA, August 2001.
    Google ScholarLocate open access versionFindings
  • J. Yin, et al. Separating Agreement from execution for Byzantine Fault Tolerant Services, in Proceedings of the 19th ACM Symposium on Operating Systems Principles, Octobor 2003.
    Google ScholarLocate open access versionFindings
  • L. Lamport, R. Shostak and M. Pease, The Byzantine Generals Problem, ACM Transactions on Programming Languages and Systems, 4(3):382-401, July 1982
    Google ScholarLocate open access versionFindings
  • L. Lamport, Time, Clocks and the Ordering of Events in a Distributed System, Communications of the ACM 21, 7 (July 1978), 558-565.
    Google ScholarLocate open access versionFindings
  • M. K. Aguilera, W. Chen, and S. Toueg, Failure detection and consensus in the crash-recovery model. Distributed Computing, Springer-Verlag, 13:2, April 2000, pp. 99-125.
    Google ScholarLocate open access versionFindings
  • M. Castro, B. Liskov, Practical Byzantine Fault Tolerance. in Proceedings of the Third Symposium on Operating Systems Design and Implementation, New Orleans, February 1999.
    Google ScholarLocate open access versionFindings
  • M. G. Velazquez, A Survey of Distributed Mutual Exclusion Algorithms, Colorado State University, Technical Report CS-93-116.
    Google ScholarFindings
  • N. Abramson, The Aloha System – Another Alternative for Computer Communications. In AFIPS Conference Proceedings, Vol. 36, 1970, pp. 295-298.
    Google ScholarLocate open access versionFindings
  • S. Saroiu, P. Krishna Gummadi, S. D. Gribble, A Measurement Study of Peer-to-Peer File Sharing Systems, in Proceedings of Multimedia Computing and Networking (MMCN) 2002.
    Google ScholarLocate open access versionFindings
  • S. Ratnasamy, P. Francis, M. Handley, R. Karp, and S. Shenker, A Scalable Content-Addressable Network, in Proceedings of ACM SIGCOMM 2001.
    Google ScholarLocate open access versionFindings
  • S. Lin, Q. Lian, M. Chen, and Z. Zhang, A Practical Distributed Mutual Exclusion Protocol in Dynamic Peer-to-Peer Systems, Microsoft Research, Technical Report MSR-TR-2004-13.
    Google ScholarLocate open access versionFindings
  • T. D. Chandra and S. Toueg, Unreliable failure detectors for reliable distributed systems. Journal of the ACM, 43(2):255-267.
    Google ScholarLocate open access versionFindings
  • Z. Zhang, The Power of DHT as a Logical Space. To appear in FTDCS’04.Distrbuted Computing(FTDCS’04)
    Google ScholarFindings
  • W. Chen, S. Lin, Q. Lian and Z. Zhang. Sigma: A Fast Fault-Tolerant Mutual Exclusion Algorithm in Dynamic Distributed System. Submission in preparation.
    Google ScholarFindings
0
Your rating :

No Ratings

Tags
Comments
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn