Martin wrote:
[...]
> Indeed, is there *any* need for testing the schedule any more often than 
> min(TSI, minimum WU deadline, 0.75*(cache length time) ) ?

SPECIAL NOTE TO JM7

Note, this is fully understood to be an INCOMPLETE FIX.

There is ALSO the problem of what tests are needed and what resultant 
rescheduling is done (if any).

Unless that is the tests truly are of ZERO cost... Good is to avoid the 
"if any" cases before we do get an exponential explosion in the present 
trigger events.


Can we have BOTH "correct tests and rescheduling" and ALSO test triggers 
that only hit the scheduler when some action is 'likely' to be needed?

No need for the triggers to 'cry wolf'... :-)


Regards,
Martin

-- 
--------------------
Martin Lomas
m_boincdev ml1 co uk.ddSPAM.dd
--------------------
_______________________________________________
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.

Reply via email to