You are viewing a plain text version of this content. The canonical link for it is here.

Posted to reviews@kudu.apache.org by "zhen.zhang (Code Review)" <ge...@cloudera.org> on 2017/12/21 12:14:22 UTC

[kudu-CR] KUDU-2238. DMS not flush under memory pressure.

zhen.zhang has uploaded this change for review. ( http://gerrit.cloudera.org:8080/8904


Change subject: KUDU-2238. DMS not flush under memory pressure.
......................................................................

KUDU-2238. DMS not flush under memory pressure.

When we choose DMS to flush, now we always pick the DMS with
highest log retention. However, as KUDU-2238 shows, in some cases
DMS with highest log retention may only consume little memory, and
other DMSs may consume much more memory, but get no change to be
flushed, even under memory pressure.

This patch gives the ability to take memory consumption into
consideration when we choose DMS.

Change-Id: I0c04d76aa0e3888352dc56eeb493a5437ef47e42
---
M src/kudu/tablet/tablet.cc
M src/kudu/tablet/tablet.h
M src/kudu/tablet/tablet_replica_mm_ops.cc
3 files changed, 19 insertions(+), 8 deletions(-)



  git pull ssh://gerrit.cloudera.org:29418/kudu refs/changes/04/8904/1
-- 
To view, visit http://gerrit.cloudera.org:8080/8904
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: newchange
Gerrit-Change-Id: I0c04d76aa0e3888352dc56eeb493a5437ef47e42
Gerrit-Change-Number: 8904
Gerrit-PatchSet: 1
Gerrit-Owner: zhen.zhang <zh...@xiaomi.com>

[kudu-CR] KUDU-2238. DMS not flush under memory pressure.

Posted by "Todd Lipcon (Code Review)" <ge...@cloudera.org>.

Todd Lipcon has posted comments on this change. ( http://gerrit.cloudera.org:8080/8904 )

Change subject: KUDU-2238. DMS not flush under memory pressure.
......................................................................


Patch Set 1:

(1 comment)

http://gerrit.cloudera.org:8080/#/c/8904/1/src/kudu/tablet/tablet.cc
File src/kudu/tablet/tablet.cc:

http://gerrit.cloudera.org:8080/#/c/8904/1/src/kudu/tablet/tablet.cc@1897
PS1, Line 1897:   double mem_weight = FLAGS_mem_weight_when_choose_best_dms_to_flush;
instead of a new parameter maybe we should be basing this on whether the system is under memory pressure (eg using process_memory::UnderMemoryPressure)? ie if memory is tight we should prioritize on size, and if not, we should prioritize on retention? Then we wouldn't need the new tunable (which I'm not sure how to tune well)



-- 
To view, visit http://gerrit.cloudera.org:8080/8904
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0c04d76aa0e3888352dc56eeb493a5437ef47e42
Gerrit-Change-Number: 8904
Gerrit-PatchSet: 1
Gerrit-Owner: zhen.zhang <zh...@xiaomi.com>
Gerrit-Reviewer: Jean-Daniel Cryans <jd...@apache.org>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Mike Percy <mp...@apache.org>
Gerrit-Reviewer: Todd Lipcon <to...@apache.org>
Gerrit-Comment-Date: Thu, 21 Dec 2017 17:59:26 +0000
Gerrit-HasComments: Yes