Supervisor Recovery in Ring Networks
Original Publication Date: 1985-Jan-01
Included in the Prior Art Database: 2005-Feb-18
Many local area ring networks employ either a monitor, a synchronizer, or a controller to ensure proper operation of the network. Unfortunately, these supervisors are also subject to failure. Therefore, designers of such networks have devised mechanisms so that every node on the network is a potential supervisor. When the active supervisor fails, other nodes acting as secondary supervisors execute a distributed election algorithm to select one of them as the next active supervisor. Some prior algorithms elect a supervisor in order n time, where n is the number of nodes participating in the election process. In this disclosure we present a new algorithm for election in a local area ring network.