> [email protected] wrote: >> Paul: >> >> The events that drive rr_sim: > > OK, so trying a little KISS: > >> RPC complete: We have now committed to some more work. Is there >> anything >> that needs to start right now to complete on time. > > Nope. No need. The request should be reasonable enough and the WUs > committed to should be reasonable enough that there should be no change > to the presently running work. > >> File Download Complete: We now have a task ready to run. Does it need >> to >> get started right now? > > Again, should be no change for the above reasoning.
Sorry, slight disagreement with Martin here. 1) RPC complete. Tasks are quantised. The client can request 1 second, but it will receive more. There are reported cases (rare, and as yet not explained) when a perfectly reasonable small request is responded to with a great deluge of work (http://setiathome.berkeley.edu/forum_thread.php?id=51410#853921). One extra test iteration - perhaps skipped if no work is allocated - seems a small price to pay for a pair of braces (suspenders) to reinforce the belt supplied by the server. 2) Download complete As somebody pointed out a long time ago in this thread, owing to bandwidth congestion or whatever, a file download may complete many hours or even days after it was initiated - and it may need to run immediately. We may know that already if one of the other triggers fired, but we may not - and in either event, we need to trigger at least an 'enforce' test post-download. _______________________________________________ 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.
