d1681e
From b6cf34d99a8b60bcf07b557328d7d998a71653ac Mon Sep 17 00:00:00 2001
d1681e
From: Prasanna Kumar Kalever <prasanna.kalever@redhat.com>
d1681e
Date: Fri, 20 Apr 2018 17:26:12 +0530
d1681e
Subject: [PATCH 303/305] block-profile: enable cluster.eager-lock in
d1681e
 block-profile
d1681e
d1681e
Eager-lock gave 2.5X perf improvement. On top of that with batching
d1681e
fix in tcmu-runner and client-io-threads we are seeing close to 3X perf
d1681e
improvement. But we don't want to include that in the default profile
d1681e
option but enable it on a case by case basis. So not adding
d1681e
client-io-threads option.
d1681e
d1681e
 >BUG: 1573119
d1681e
 >Upstream-patch: https://review.gluster.org/19913
d1681e
BUG: 1569951
d1681e
Change-Id: Ida53c3ef9a041a73b65fdd06158ac082da437206
d1681e
Signed-off-by: Prasanna Kumar Kalever <prasanna.kalever@redhat.com>
d1681e
Reviewed-on: https://code.engineering.redhat.com/gerrit/141310
d1681e
Tested-by: RHGS Build Bot <nigelb@redhat.com>
d1681e
Reviewed-by: Sunil Kumar Heggodu Gopala Acharya <sheggodu@redhat.com>
d1681e
---
d1681e
 extras/group-gluster-block | 2 +-
d1681e
 1 file changed, 1 insertion(+), 1 deletion(-)
d1681e
d1681e
diff --git a/extras/group-gluster-block b/extras/group-gluster-block
d1681e
index d619255..56b406e 100644
d1681e
--- a/extras/group-gluster-block
d1681e
+++ b/extras/group-gluster-block
d1681e
@@ -6,7 +6,7 @@ performance.open-behind=off
d1681e
 performance.readdir-ahead=off
d1681e
 performance.strict-o-direct=on
d1681e
 network.remote-dio=disable
d1681e
-cluster.eager-lock=disable
d1681e
+cluster.eager-lock=enable
d1681e
 cluster.quorum-type=auto
d1681e
 cluster.data-self-heal-algorithm=full
d1681e
 cluster.locking-scheme=granular
d1681e
-- 
d1681e
1.8.3.1
d1681e