[1]李传佑,汪芸.拜占庭环境下新成员加入容错组状态同步[J].东南大学学报(自然科学版),2010,40(1):23-28.[doi:10.3969/j.issn.1001-0505.2010.01.005]
 Li Chuanyou,Wang Yun.State synchronization for new member join in Byzantine-tolerant environments[J].Journal of Southeast University (Natural Science Edition),2010,40(1):23-28.[doi:10.3969/j.issn.1001-0505.2010.01.005]
点击复制

拜占庭环境下新成员加入容错组状态同步()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
40
期数:
2010年第1期
页码:
23-28
栏目:
计算机科学与工程
出版日期:
2010-01-20

文章信息/Info

Title:
State synchronization for new member join in Byzantine-tolerant environments
作者:
李传佑 汪芸
东南大学计算机科学与工程学院,南京 210096; 东南大学网络和信息集成教育部重点实验室,南京 210096
Author(s):
Li Chuanyou Wang Yun
School of Computer Science and Engineering, Southeast University, Nanjing 210096, China
Key Laboratory of Computer Network and Information Integration of Ministry of Education, Southeast University, Nanjing 210096, China
关键词:
Erasure Coding 拜占庭错误 状态同步
Keywords:
Erasure Coding Byzantine fault state synchronization
分类号:
TP393.08
DOI:
10.3969/j.issn.1001-0505.2010.01.005
摘要:
在主动复制技术下,为了容忍少量节点的拜占庭错误并提高组成员加入时状态同步的效率,提出了快速状态同步协议FSSP.FSSP利用Erasure Coding将状态数据分成P块,经过线性运算,编码成Q(Q>P).新加入节点只需获得Q块中的任意P块数据即可完成解码,获得状态数据.同时FSSP使用Hash技术屏蔽了拜占庭节点带来干扰.仿真实验结果表明:在100 Mbit/s以太网环境下,网络传输时延是系统的主要瓶颈,无论待同步状态数据驻留在内存还是硬盘中,FSSP均要优于直接同步协议DSSP.这是因为FSSP有效地减少了网络中传输的报文量,以少量的编解码计算代价换取了较大的网络传输时延,最终达到了加快状态同步过程的目的.
Abstract:
With active replication technique, in order to tolerate Byzantine failure that a few nodes may suffer from and improve the performance of state synchronization when new member joins in, a novel fast state synchronization protocol(FSSP)is proposed. By utilization of Erasure Coding, FSSP divides the state data into P blocks which are then encoded into Q blocks(Q>P)through linear operation. The new member only needs to get P blocks among Q to decode the original state data. Meanwhile, FSSP can mask some Byzantine failure by using Hash technique. The simulation results show that with 100 Mbit/s Ethernet, the delay of network transmission is the bottleneck. FSSP is better than directly synchronization protocol(DSSP)no matter whether the state data is in memory or in disk. Although FSSP brings a little cost of coding and decoding, it reduces a lot of data transferring in networks and finally it achieves the purpose of improving the performance of state synchronization.

参考文献/References:

[1] Guerraoui R,Schiper A.Software-based replication for fault tolerance [J].IEEE Computer,1997,30(4):68-74.
[2] Guerraoui R,Schiper A.Fault-tolerance by replication in distributed systems [J].Lecture Notes in Computer Science,1996,1088:38-57.
[3] Object Management Group.Fault tolerant CORBA,common object request broker architecture,V.3.0[S].Massachusetts,USA:OMG,2002.
[4] Goodson G R,Wylie J J,Ganger G R,et al.Efficient byzantine-tolerant erasure-coded storage[C] //Proc of Dependable Systems and Networks(DSN).Florence,Italy,2004:135-144.
[5] Lamport L,Shostak R,Pease M.The byzantine generals problem [J].ACM Transactions on Programming Languages and Systems,1982,4(3):382-401.
[6] Castro M,Liskov B.Practical byzantine fault tolerance [C] //Proc of OSDI.New Orleans,USA,1999:173-186.
[7] Kotla R,Alvisi L,Dahlin M,et al.Zyzzyva:speculative byzantine fault tolerance[C] //Proc of SOSP.Stevenson,USA,2007:45-58.
[8] Malkhi D,Reiter M.Byzantine quorum systems [J].Distributed Computing,1998,11(4):569-578.
[9] Abd-El-Malek M,Ganger G,Goodson G,et al.Fault-scalable byzantine fault-tolerant services [C] //Proc of SOSP.Brighton,UK,2005:59-74.
[10] Cowling J,Myers D,Liskov B,et al.HQ replication:a hybrid quorum protocol for byzantine fault tolerance [C] //Proc of OSDI.Seattle,USA,2006:177-190.
[11] Kihlstrom K P,Moser L E,Melliar-Smith P M.The securering group communication system [J]. ACM Transactions on Information and System Security,2001,4(4):371-406.
[12] Narashimhan P,Moser L E,Melliar-Smith P M.State synchronization and recovery for strongly consistent replicated CORBA objects [C] //Proc of Dependable Systems and Networks(DSN).Göteborg,Sweden,2001:261-270.
[13] Wang Y,Wang J L.Non-blocking message total ordering protocol [J]. Science in China Series F:Information Science,2008,51(12):1919-1934.
[14] Dimakis A G,Prabhakaran V,Ramchandran K.Decentralized erasure codes for distributed networked storage [J]. IEEE Transactions on Information Theory,2006,52(6):2809-2816.
[15] Plank J S,Xu L.Optimizing cauchy reed-solomon codes for fault-tolerant network storage applications.[C] //Proc of 5th IEEE International Symposium on Network Computing and Application(NCA).Massachusetts,USA,2006:173-180.
[16] Dimakis A G,Godfrey P B,Wainwright M J,et al.Network coding for distributed storage systems[C] //Proc of the IEEE INFOCOM.Alaska,USA,2007:2000-2008.
[17] Plank J S.A tutorial on Reed-Solomon coding for fault-tolerance in RAID-like systems [J]. Software Practice & Experience,1997,27(9):995-1012.
[18] Young Eric A,Hudson Tim J.Open SSLo.98i[EB/OL].(2008-09-15)[2009-06-20].http://www.openssl.org/.

备注/Memo

备注/Memo:
作者简介: 李传佑(1985—),男,博士生; 汪芸(联系人),女,博士,教授,博士生导师,yunwang@seu.edu.cn.
基金项目: 国家自然科学基金资助项目(60793122)、国家重点基础研究发展计划(973计划)资助项目(2009CB320705).
引文格式: 李传佑,汪芸.拜占庭环境下新成员加入容错组状态同步[J].东南大学学报:自然科学版,2010,40(1):23-28. [doi:10.3969/j.issn.1001-0505.2010.01.005]
更新日期/Last Update: 2010-01-20