Electing "Good" Leaders
Electing "Good" Leaders
复制标题
选出“好”领导人
DOI：
--
复制
发表时间：
1994
期刊：
影响因子：
--
通讯作者：
J. Kurose
中科院分区：
文献类型：
--
作者：
Suresh Singh;J. Kurose
在分布式系统中，用于选择杰出节点或领导者来协调某些活动的算法称为领导者选举算法。虽然基于节点 ID 号的传统领导者选举方法具有相当大的理论意义，但我们认为这些方法不太实用，因为选出的领导者不一定会提供良好的性能。在本文中，我们提出并研究了受社会启发的领导者选举方案，该方案试图将领导者定位在系统中的“好”节点（从性能角度来看）。每个节点使用本地可用的信息来投票给各个候选人。使用选举方案将这些选票结合起来以确定领导者。我们提出和检查的选举方案被证明几乎与传统的基于优化的领导者选举方法一样好，并且对节点故障具有鲁棒性。
In a distributed system, an algorithm used to select a distinguished node or leader to coordinate some activity is known as a leader election algorithm. While traditional approaches to leader election, based upon node ID numbers, are of considerable theoretical interest, we believe these approaches are not very practical since the leader elected will not necessarily deliver good performance. In this paper, we propose and examine socially inspired leader election schemes that attempt to locate the leader at a "good" node (from a performance standpoint) in the system. Each node uses locally available information to vote for the various candidates. These votes are combined using an election scheme to determine the leader. The election schemes we propose and examine are shown to perform almost as well as a traditional optimization-based approach toward leader election and are shown to be robust to node failures.