On Thursday 28 June 2007 17:02, Lindsay Haisley wrote: > I don't know what would be gained by a random wait.
The idea of a random wait for contention resolution is long standing. It's built into the TCP/IP protocol for example. For example, say my cron job runs at 3 am. Lot's of them probably do. This causes the congestion. Waiting a random time makes the peaks gradually level out on the second retry. -- Phil Barnett AI4OF SKCC #600
