Thanks for the information. -Sreenath ------------------------- Date: Wed, 25 Mar 2015 04:11:11 +0100 From: Francois Lafont <flafdivers@xxxxxxx> To: ceph-users <ceph-users@xxxxxxxx> Subject: Re: PG calculator queries Message-ID: <5512274F.1000003@xxxxxxx> Content-Type: text/plain; charset=utf-8 Hi, Sreenath BH wrote : > consider following values for a pool: > > Size = 3 > OSDs = 400 > %Data = 100 > Target PGs per OSD = 200 (This is default) > > The PG calculator generates number of PGs for this pool as : 32768. > > Questions: > > 1. The Ceph documentation recommends around 100 PGs/OSD, whereas the > calculator takes 200 as default value. Are there any changes in the > recommended value of PGs/OSD? Not really I think. Here http://ceph.com/pgcalc/, we can read: Target PGs per OSD This value should be populated based on the following guidance: - 100 If the cluster OSD count is not expected to increase in the foreseeable future. - 200 If the cluster OSD count is expected to increase (up to double the size) in the foreseeable future. - 300 If the cluster OSD count is expected to increase between 2x and 3x in the foreseeable future. So, it seems to me cautious to recommend 100 in the official documentation because you can increase the pg_num but it's impossible to decrease it. So, if I should recommend just one value, It would be 100. > 2. Under "notes" it says: > "Total PG Count" below table will be the count of Primary PG copies. > However, when calculating total PGs per OSD average, you must include > all copies. > > However, the number of 200 PGs/OSD already seems to include the > primary as well as replica PGs in a OSD. Is the note a typo mistake or > am I missing something? To my mind, in the site, the "Total PG Count" doesn't include all copies. So, for me, there is no typo. Here is 2 basic examples from http://ceph.com/pgcalc/ with just *one* pool. 1. Pool-Name Size OSD# %Data Target-PGs-per-OSD Suggested-PG-count rbd 2 10 100.00% 100 512 2. Pool-Name Size OSD# %Data Target-PGs-per-OSD Suggested-PG-count rbd 2 10 100.00% 200 1024 In the first example, I have: 512/10 = 51.2 but (Size x 512)/10 = 102.4 In the second example, I have: 1024/10 = 102.4 but (Size x 1024)/10 = 204.8 HTH. -- Fran?ois Lafont _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |