|
Distributed operating systemsThus method 1 wastes bandwidth to reduce interrupts compared to method 2. Depending on the average message size, one may be preferable to the other. In summary, this protocol allows reliable broadcasting to be done on an unreliable network in just over two messages per reliable broadcast. Each broadcast is indivisible, and all applications receive all messages in the same order, no matter how many are lost. The worst that can happen is that some delay is introduced when a message is lost, which rarely happens. If two processes attempt to broadcast at the same time, one of them will get to the sequencer first and win. The other will see a broadcast from its competitor coming back from the sequencer, and will realize that its request has been queued and will appear shortly, so it simply waits. Fault-Tolerant Group Communication So far we have assumed that no processors crash. In fact, this protocol has been designed to withstand the loss of an arbitrary collection of k processors (including the sequencer), where k (the resilience degree) is selected when the group is created ...» |
Код для вставки книги в блог HTML
phpBB
текст
|
|