Consensus on transaction commit

被引:152
作者
Gray, Jim
Lamport, Leslie
机构
[1] Microsoft Res, San Francisco, CA 94105 USA
[2] Microsoft Res, Mountain View, CA 94043 USA
来源
ACM TRANSACTIONS ON DATABASE SYSTEMS | 2006年 / 31卷 / 01期
关键词
algorithms; reliability; consensus; Paxos; two-phase commit;
D O I
10.1145/1132863.1132867
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The distributed transaction commit problem requires reaching agreement on whether a transaction is committed or aborted. The classic Two-Phase Commit protocol blocks if the coordinator fails. Fault-tolerant consensus algorithms also reach agreement, but do not block whenever any majority of the processes are working. The Paxos Commit algorithm runs a Paxos consensus algorithm on the commit/abort decision of each participant to obtain a transaction commit protocol that uses 2F + 1 coordinators and makes progress if at least F + 1 of them are working properly. Paxos Commit has the same stable-storage write delay, and can be implemented to have the same message delay in the fault-free case as Two-Phase Commit, but it uses more messages. The classic Two-Phase Commit algorithm is obtained as the special F = 0 case of the Paxos Commit algorithm.
引用
收藏
页码:133 / 160
页数:28
相关论文
共 19 条
  • [11] GUERRAOUI R, 1995, LNCS, V972, P87
  • [12] Lamport L., 2001, SIGACT News, V32, P51
  • [13] The part-time parliament
    Lamport, L
    [J]. ACM TRANSACTIONS ON COMPUTER SYSTEMS, 1998, 16 (02): : 133 - 169
  • [14] Lamport L., 2003, SPECIFYING SYSTEMS
  • [15] LAMPSON BW, 1996, LECT NOTES COMPUTER, V1151, P1
  • [16] MOHAN C, 1983, P 2 ANN ACM S PRINC, P29
  • [17] Newcomer E., 2002, UNDERSTANDING WEB SE
  • [18] REACHING AGREEMENT IN THE PRESENCE OF FAULTS
    PEASE, M
    SHOSTAK, R
    LAMPORT, L
    [J]. JOURNAL OF THE ACM, 1980, 27 (02) : 228 - 234
  • [19] SKEEN D, 1981, P ACM SIGMOD INT C M, P133, DOI [10.1145/582318.582339, DOI 10.1145/582318.582339]