Hi David,

On Tuesday 21 July 2009 22:42:47 David Anderson wrote:
> > 2) Are there any plans to move the plan_class configuration from
> > sched/sched_plan.cpp to the database for example? Storing the plan_class
> > details (e.g. min versions, min. global mem) as well as the host_usage
> > (e.g. number of CPUs) details in database tables (i.e. referenced by
> > app_version) seems to be more natural than having these things hard-coded
> > in the scheduler. This makes it much harder and error prone to release a
> > new CUDA app version because you've to change the scheduler
> > simultaneously...
>
> no such plans; putting the app-selection logic in C++ code
> rather than the database was a design decision (and a good one).

What's the rationale behind that decision? If your concern is that even more 
DB fetches might hurt the scheduler's performance one could instead use some 
kind of config file. Of course one should put the *logic* into the code but not 
necessarily the config values to tune the logic, right? I can think of a number 
of individual "cuda" plan classes that only vary slightly in their particular 
configuration (e.g. required global memory), not in their general logic.


Cheers,
Oliver

_______________________________________________
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