From: Loic Dachary Date: Fri, 23 May 2014 05:25:37 +0000 (+0200) Subject: documentation: adapt PG formula for erasure coded pools X-Git-Tag: v0.82~41^2 X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=ab0db34eaa906b7657d8dd747d15d44fbc3d310f;p=ceph.git documentation: adapt PG formula for erasure coded pools Signed-off-by: Loic Dachary --- diff --git a/doc/rados/operations/placement-groups.rst b/doc/rados/operations/placement-groups.rst index 127fae9bf53..0fa0f1212cc 100644 --- a/doc/rados/operations/placement-groups.rst +++ b/doc/rados/operations/placement-groups.rst @@ -43,9 +43,13 @@ across your cluster. We recommend approximately 50-100 placement groups per OSD to balance out memory and CPU requirements and per-OSD load. For a single pool of objects, you can use the following formula:: - (OSDs * 100) - Total PGs = ------------ - Replicas + (OSDs * 100) + Total PGs = ------------ + OSD per object + +Where **OSD per object** is the number of replicas (or pool size) for +replicated pools or the K+M sum for erasure coded pools (as returned +by **ceph osd erasure-code-profile get**). The result should be **rounded up to the nearest power of two.** Rounding up is optional, but recommended if you want to ensure that