summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorManikandan Selvaganesh <mselvaga@redhat.com>2016-04-04 11:07:31 +0530
committerNiels de Vos <ndevos@redhat.com>2016-04-12 03:18:15 -0700
commitd05734c3b80f439edc4b423f827121eea0a2cf95 (patch)
treea75e02c4b0f0cc92cae4473e1cd30da268e48885
parent097e8cee9b439d2c56f89fe3868e8896c99ff184 (diff)
quota: enhance quota enable/disable process
Adding design document for enhancing quota enable/disable process Change-Id: Ibc2b6816d5a9ea80ac5c0b921122436bfc347503 Signed-off-by: Manikandan Selvaganesh <mselvaga@redhat.com> Reviewed-on: http://review.gluster.org/13897 Reviewed-by: Niels de Vos <ndevos@redhat.com> Tested-by: Niels de Vos <ndevos@redhat.com>
-rw-r--r--under_review/quota-enhancements.md77
1 files changed, 77 insertions, 0 deletions
diff --git a/under_review/quota-enhancements.md b/under_review/quota-enhancements.md
new file mode 100644
index 0000000..a5412a4
--- /dev/null
+++ b/under_review/quota-enhancements.md
@@ -0,0 +1,77 @@
+Feature
+-------
+Quota Enhancements for Gluster
+
+Summary
+-------
+Enhance Quota enable/disable process.
+
+Owners
+------
+Vijaikumar Mallikarjuna <vmallika@redhat.com>
+Manikandan Selvaganesh <mselvaga@redhat.com>
+
+Current status
+--------------
+Under review
+
+Related Feature Requests and Bugs
+---------------------------------
+Currently, Quota crawl process is being done by a single mount point. This
+process is very slow if there are huge number of files.
+
+https://bugzilla.redhat.com/show_bug.cgi?id=1290766
+
+Detailed Description
+--------------------
+Previously, when quota is enabled or disabled on a volume, crawl proces is
+done from a single mount point to create/remove the quota related extended
+attributes. When there are huge number of files that exists in the volume,
+this process tends to be very slow taking a long time depending on the number
+of files.
+
+The proposed feature will spawn the crawl process for each brick in the volume
+and files will be checked in parallel which is an independent process for every
+brick. This improves the speed of crawling process, thus enhancing the quota
+enable/disable process.
+
+Benefit to GlusterFS
+--------------------
+Quota enable/disable process is made faster.
+
+Scope
+-----
+
+#### Nature of proposed change
+Changes are in glusterd code path where crawling process is initiated.
+
+#### Implications on manageability
+None
+
+#### Implications on presentation layer
+None
+
+#### Implications on persistence layer
+None
+
+#### Implications on 'GlusterFS' backend
+None
+
+#### Modification to GlusterFS metadata
+None
+
+#### Implications on 'glusterd'
+None
+
+How To Test
+-----------
+When quota is enabled or disabled, the crawl process will be created for each
+brick.
+
+Dependencies
+------------
+None
+
+Comments and Discussion
+-----------------------
+