System and method for adjusting membership of a data replication group

作者: Timothy Andrew Rath , Jakub Kulesza , David Alan Lutz

DOI:

关键词:

摘要: A system that implements a data storage service may store on behalf of clients. The maintain in multiple replicas partitions are stored respective computing nodes the system. master replica for group increment membership version indicator group, and propagate metadata (including indicator) indicating change to other members group. Propagating include sending log record containing be appended their logs. Once becomes durable, it committed. attempting become determine another has observed more recent version, which case logs synchronized or snipped, attempt abandoned.

参考文章(63)
Avishai Wool, Quorum Systems in Replicated Databases: Science or Fiction? IEEE Data(base) Engineering Bulletin. ,vol. 21, pp. 3- 11 ,(1998)
Nathan Lorenzo Binkert, Indrajit Roy, Mehul A. Shah, Wojciech Golab, Bruce Walker, Server replication and transaction commitment ,(2011)
Bugra Gedik, Kun-Lung Wu, Henrique Andrade, Vibhore Kumar, Louis R. Degenaro, Gabriels Jacques da Silva, High availability operator groupings for stream processing applications ,(2009)
Vishal Kher, Craig Harmer, Prasanta Dash, Locating the latest version of replicated data files ,(2009)
David Chess, Murthy Devarakonda, William Arnold, Alla Segal, Ian Whalley, Policy-based, autonomically allocated storage ,(2003)
Rod Gamache, Sunita Shrivastava, John D. Vert, Method and system for replicated and consistent modifications in a server cluster ,(1998)
Michael Shannon Irizarry, Claudio Taglienti, Jaideep Abichandani, Narothum Saxena, Andrew Sofronas, Service utilization control manager ,(2007)