PG count isn’t just about storage size, it also affects performance, parallelism, and recovery.
You want pgp_num for RBD metadata pool to be at the VERY least the number of OSDs it lives on, rounded up to the next power of 2. I’d probably go for at least (2x#OSD) rounded up. If you have two few, your metadata operations will contend with each other. > On Nov 3, 2022, at 10:24, mailing-lists <[email protected]> wrote: > > Dear Ceph'ers, > > I am wondering on how to choose the number of PGs for a RBD-EC-Pool. > > To be able to use RBD-Images on a EC-Pool, it needs to have an regular > RBD-replicated-pool, as well as an EC-Pool with EC overwrites enabled, but > how many PGs would you need for the RBD-replicated-pool. It doesn't seem to > eat a lot of storage, so if I'm not mistaken, it could be actually a quite > low number of PGs, but is this recommended? Is there a best practice for this? > > > Best > > Ken > > _______________________________________________ > ceph-users mailing list -- [email protected] > To unsubscribe send an email to [email protected] _______________________________________________ ceph-users mailing list -- [email protected] To unsubscribe send an email to [email protected]
