I've read through this a couple of times, and I think it's the right way to go.
It takes the burden off the project having to tune credits and FLOPs, and I think that's where the whole issue starts: once the project goes public, and the results are flowing, they don't see much of a need to tune -- especially if the project staff is not from a Computer Science background. My only question is: if credit is to be granted on a fixed credit per workunit, independent of time, will we have people aborting the long work units in favor of the short ones? Even if all the estimates were the same on SETI, it'd be possible to recognize "slow" workunits from the angle-range and abort them. -- Lynn David Anderson wrote: > I'm coming around to the viewpoint that projects shouldn't be expected > to supply estimates of job duration or application performance. > I think it's feasible to maintain these estimates dynamically, > based on actual job runtimes. > I've sketched a set of changes that would accomplish this: > http://boinc.berkeley.edu/trac/wiki/AutoFlops > Comments welcome. > > BTW, a bonus of the proposed design is that it provides > a project-independent credit-granting policy. > > -- David > > Richard Haselgrove wrote: >> ... if projects >> are expected to fine-tune performance metrics down to the individual >> plan_class level, then I'm sorry, but they just won't. I've had to shout >> (loudly and repeatedly) at both AQUA and GPUGrid to get them to adjust >> rsc_fpops_est to within an order of magnitude of reality (in AQUA's case, >> two orders of magnitude). > _______________________________________________ > 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. > _______________________________________________ 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.
