keith-turner opened a new pull request, #5728:
URL: https://github.com/apache/accumulo/pull/5728

   Each fate thread that was looking for a single fate operation to work on 
would read all fate txs from zookeeper.  This does N^2 work and would place a 
lot of load on zookeeper and the manager as the number of total fate 
transaction increased.  This change avoids reading everything from zookeeper 
when looking for a single thing to work on.
   
   Ran the random walk bulk test w/ this change and saw a much lower CPU usage 
for the zookeeper server.  Also ran `mvn verify -Psunny` successfully against 
these changes.
   
   fixes #5712


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to