grwilson commented on this pull request.
> @@ -1039,6 +1039,13 @@ metaslab_group_allocatable(metaslab_group_t *mg, > metaslab_group_t *rotor, if (mg->mg_no_free_space) return (B_FALSE); + /* + * Relax allocation throttling for ditto blocks. Due to + * random imbalances in allocation it tends to push copies + * to one vdev, that looks a bit better at the moment. + */ + qmax = qmax * (4 + d) / 4; Out of curiosity, did you find that it made a difference to increase the qmax value between the 2nd and 3rd DVAs? Or is equally efficient to say any DVA > 1 gets to go over the qmax value by 25%? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/openzfs/openzfs/pull/688#discussion_r221943428 ------------------------------------------ openzfs: openzfs-developer Permalink: https://openzfs.topicbox.com/groups/developer/T5b2a8092d211827d-M6fd39e132f4291f80b4382a6 Delivery options: https://openzfs.topicbox.com/groups/developer/subscription