somandal opened a new pull request, #15001:
URL: https://github.com/apache/pinot/pull/15001

   Today the config: 
`pinot.server.instance.table.level.max.parallel.segment.downloads` is set up 
based on the server configuration. This PR aims to make this config dynamically 
configurable via the ZK cluster configs so that a server restart is not 
required to pick up changes to this configuration.
   
   Another change this PR makes is to always create the semaphore to control 
the parallelism and removes the behavior when this configuration is set to `-1` 
to indicate downloads can be unlimited. Instead it sets a high default value 
for parallel downloads instead. To handle backward compatibility, in case the 
server configuration sets this to `-1` the init() code will default it to a 
high default value instead. ZK cluster config updates must set it to be > 0 
though.
   
   The above should be okay as the maximum threads that could be downloading 
segments at any given time will be based on the value of 
STATE_TRANSITIONS.maxThreads (which is 40 by default) + 
max.parallel.refresh.threads
   
   cc @Jackie-Jiang @klsince @npawar @jackjlli @jasperjiaguo 


-- 
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: commits-unsubscr...@pinot.apache.org

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@pinot.apache.org
For additional commands, e-mail: commits-h...@pinot.apache.org

Reply via email to