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

Posted to dev@zookeeper.apache.org by "Fangmin Lv (JIRA)" <ji...@apache.org> on 2018/08/09 04:40:00 UTC

[jira] [Created] (ZOOKEEPER-3116) Make the DataTree.approximateDataSize more efficient

Fangmin Lv created ZOOKEEPER-3116:
-------------------------------------

             Summary: Make the DataTree.approximateDataSize more efficient
                 Key: ZOOKEEPER-3116
                 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-3116
             Project: ZooKeeper
          Issue Type: Improvement
          Components: server
            Reporter: Fangmin Lv
            Assignee: Fangmin Lv
             Fix For: 3.6.0


The approximateDataSize is a nice metric to show what's the total size stored in ZooKeeper ensemble over time, but it's expensive to query too often, since each query will go through all the nodes to calculate the total size.

It's better to use a counter to record the total data size when txns applied to the DataTree, which is cheaper.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)