O'Reilly logo

Apache Accumulo for Developers by Guðmundur Jón Halldórsson

Stay ahead with the world's most comprehensive technology and business learning platform.

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, tutorials, and more.

Start Free Trial

No credit card required

ZooKeeper performance

To be able to understand and spot problems in ZooKeeper cluster, we need to establish a baseline. On the ZooKeeper website is the Service Latency Overview page which is a good starting ground to establish the baseline:

http://wiki.apache.org/hadoop/ZooKeeper/ServiceLatencyOverview

Because every node in ZooKeeper has to vote, it will cause an increased cost in writing operations when the cluster size increases. The flipside is, if too many clients connect to the same node, then it's possible for that node to go down. Then you will only have three ZooKeeper servers running (1 leader and 2 followers), and after one is down, only one follower will still be alive, and if all the clients connect to it, they might take it down. ...

With Safari, you learn the way you learn best. Get unlimited access to videos, live online training, learning paths, books, interactive tutorials, and more.

Start Free Trial

No credit card required