You can pass group names with the -K argument, when you start the process. When you queue a task, you must pass in the group_name parameter to "reserve" that task for a specific worker.
BTW: did you at least tried to read http://web2py.com/books/default/chapter/29/04/the-core#web2py-Scheduler before asking ? On Tuesday, May 20, 2014 9:46:39 PM UTC+2, DeanK wrote: > > I'm looking to use the built in scheduler to farm out work, but run gpu > dependent code. Because of this I need to be able to control on which node > the task is scheduled. I thought I could do this using "groups" but I > don't see how since you set the group in the scheduler constructor > according to the docs? I was expecting you would set groups that workers > are "subscribed" to when you start them on each node and then when you > schedule a task you can set which group the task is assigned to. Is this > not possible or have I misinterpreted the documentation? > > scheduler.py in models: > > def my_task1(): > #my first task > > def my_task2(): > #another task > > from gluon.scheduler import Scheduler > > db_sched = > DAL('mysql://XXXX:XXXXX@XXXXX/XXXX',pool_size=8,check_reserved=['mysql'],migrate=True,fake_migrate_all=False,lazy_tables=False) > scheduler = Scheduler(db_sched,heartbeat=2) > > > > -- Resources: - http://web2py.com - http://web2py.com/book (Documentation) - http://github.com/web2py/web2py (Source code) - https://code.google.com/p/web2py/issues/list (Report Issues) --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
