期刊名称：Bulletin of the Technical Committee on Data Engineering
出版年度：2017
卷号：40
期号：4
页码：3
出版社：IEEE Computer Society
摘要：Global-scale data management (GSDM) empowers systems by providing higher levels of fault-tolerance,read availability, and efficiency in utilizing cloud resources. This has led to the emergence of global-scale data management and event processing. However, the Wide-Area Network (WAN) latency sep-arating datacenters is orders of magnitude larger than typical network latencies, and this requires areevaluation of many of the traditional design trade-offs of data management systems. Therefore, datamanagement problems must be revisited to account for the new design space. In this paper, we revisit theproblem of supporting strongly-consistent transaction processing for GSDM. This includes providing anunderstanding of the limits imposed by the WAN latency on transaction latency in addition to a designof a system framework that aims to reduce response time and increase scalability. This infrastructureincludes a transaction processing component, a fault-tolerance component, a communication compo-nent, and a placement component. Finally, we discuss the current challenges and future directions oftransaction processing in GSDM.
