jnturton commented on issue #2978: URL: https://github.com/apache/drill/issues/2978#issuecomment-2869360284
@rymarm this looks like a great idea to me. > Is it appropriate to store information about the current number of user connections in Zookeeper? > Should client connection load balancing be enabled be configurable and enabled by default? > Is it beneficial to distribute all connections evenly across the entire Drill cluster? My own answers to all three of these are yes. > For instance, let's say you have a lot of short running queries. Would it matter if they are all going to a single Drill node? It's still possible to randomise over tied winners after sorting the list of Drillbits in order of open connection count (if there are 10 Drillbits with zero open connections, choose one of those at random). > Meaning is Drill sending the query to the node closest to the data? For storage that supports this it happens later on, during exeuction, no matter which Drillbit the client originally connected to. -- 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: dev-unsubscr...@drill.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
