Virtual Synchrony
Goal: Combine atomic multicast with failure detection and group membership. The software framework for virtual synchrony gives everyone the same view of live group members, ordered message delivery, and atomic multicasting of messages to all of these members. We often deploy distributed systems in an attempt to achieve both high availability and high scalability. High availability refers to the fraction of time that the system is up and running and capable of doing whatever it is supposed to do. High scalability refers to the ability to expand the system to handle higher amounts of traffic. Both of these goals are often achieved via redundancy: adding replicated components to the system. Redundant components can step in when some components break, thus helping with availability. Active-passive replication means that the redundant components are generally not servicing requests but are ready to take over for a failed component. They will get updates, possibly asynchronously, but will no
Goal: Combine atomic multicast with failure detection and group membership. The software framework for virtual synchrony gives everyone the same view of live group members, ordered message delivery, and atomic multicasting of messages to all of these members. We often deploy distributed systems in an attempt to achieve both high availability and high scalability. High availability refers to the fraction of time that the system is up and running and capable of doing whatever it is supposed to do. High scalability refers to the ability to expand the system to handle higher amounts of traffic. Bo
Explore this link on the map →