作者: Hiroaki Honda , Shigenari Nakamura , Hiroki Nakayama , Dilawaer Duolikun , Tomoya Enokidoy
关键词:
摘要: In distributed applications, a group of multiple processes are cooperating with one another by exchanging messages in networks. Information systems like cloud computing getting scalable, which composed huge number processes. We assume physical clock each process reads is synchronized time sever, i.e. maximum offset to UTC bounded some value. Furthermore, interconnected various types this paper, we consider scalable subgroups. Processes subgroup local network (LN) and pair subgroups global (GN). supports every reliable synchronous communication. That is, no message lost delay between nodes An LN GN support shorter longer than the offset, respectively. Messages causally delivered taking usage both linear clock. clock, not required be may unnecessarily ordered. order reduce pairs ordered additionally used. evaluate protocol show can reduced compared protocol.