Alexey Serbin has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/20310 )

Change subject: KUDU-3497 optimize leader rebalancer algorithm
......................................................................


Patch Set 2:

(16 comments)

http://gerrit.cloudera.org:8080/#/c/20310/2//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/20310/2//COMMIT_MSG@9
PS2, Line 9: especilally
especially


http://gerrit.cloudera.org:8080/#/c/20310/2//COMMIT_MSG@9
PS2, Line 9: rebalancer
rebalancing


http://gerrit.cloudera.org:8080/#/c/20310/2//COMMIT_MSG@10
PS2, Line 10: the tables who have less hash buckets and replication factors.
the tables having smaller number of hash partitions and replication factor.


http://gerrit.cloudera.org:8080/#/c/20310/2//COMMIT_MSG@33
PS2, Line 33: So the leader skew will never be more than 1.
> Please add some unit tests to prove your fix is correct.
+1

It would be nice to add a test -- that's not only to make sure the updated code 
addresses the issue described in the commit message, but also to catch possible 
regressions if the related code changes in the future.


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc
File src/kudu/master/auto_leader_rebalancer.cc:

http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@204
PS2, Line 204: remain
nit for this and the other's variable name:
  * change remain --> remaining
  * if using 'num' suffix for one variable, use it for the other as well if 
both semantically represent the number/count of some entities


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@204
PS2, Line 204: uint32_t
nit: why not to use 'auto' or 'size_t' that's the return type of the size() 
method for std containers?


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@206
PS2, Line 206: who
that


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@206
PS2, Line 206: rebalance
rebalancing


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@206
PS2, Line 206: num
number


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@209
PS2, Line 209: uint32_t replica_count = tablet_ids_ptr ? tablet_ids_ptr->size() 
: 0;
nit: is it possible to use this criterion as-is in the condition for the 'if 
()' below without introducing a variable?


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@228
PS2, Line 228: the remaining
the number of remaining


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@228
PS2, Line 228: could be divided
is divisible


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@228
PS2, Line 228: tablets
tablet replicas


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@228
PS2, Line 228: the remaining
the number of remaining


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@229
PS2, Line 229: leader num of all the remaining tablet servers
number of leader


http://gerrit.cloudera.org:8080/#/c/20310/2/src/kudu/master/auto_leader_rebalancer.cc@408
PS2, Line 408: this loop
Imagine you are a person without the context of this code and you see this info 
message in the log.  Would 'this loop' be confusing to you?

Maybe, describe what 'this loop' actually is if you want to convey some useful 
information here to the users.  Otherwise, maybe don't log anything here?  If 
you need just a message for debugging purposes, consider using VLOG(...) here 
instead.



--
To view, visit http://gerrit.cloudera.org:8080/20310
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I0f1fe796fd98da2d8764da793b7e254319e6348a
Gerrit-Change-Number: 20310
Gerrit-PatchSet: 2
Gerrit-Owner: Song Jiacheng <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Song Jiacheng <[email protected]>
Gerrit-Reviewer: Wang Xixu <[email protected]>
Gerrit-Reviewer: Yingchun Lai <[email protected]>
Gerrit-Reviewer: Yuqi Du <[email protected]>
Gerrit-Comment-Date: Tue, 08 Aug 2023 00:07:52 +0000
Gerrit-HasComments: Yes

Reply via email to