It isn't outages, it's anything that hoses up the even distribution of uploads.
Even random luck. Same is true to a lesser extent with downloads and scheduler requests. Nicolás Alvarez wrote: > El Dom 12 Jul 2009 20:58:00 Lynn W. Taylor escribió: >> If every work unit went out, was out for a few hours, and then returned, >> 24/7/365, the problem I'm talking about does not exist -- the flow is even. > > The problem I had involved no outages. A few hundred workunits went out, were > out for a few hours, then the majority tried to come back at the same time. > My project was running on a "home" Internet connection, so two or three users > could be enough to saturate the project bandwidth. > > But increase the numbers and it could happen to a "real" project too. > A project without a constant stream of work. It sends a bunch of workunits, > clients get them and go quiet for a while. If most take around the same time > to finish, you get a DDoS when they try to upload. > _______________________________________________ boinc_dev mailing list [email protected] http://lists.ssl.berkeley.edu/mailman/listinfo/boinc_dev To unsubscribe, visit the above URL and (near bottom of page) enter your email address.
