[ https://issues.apache.org/jira/browse/CASSANDRA-12015?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15334041#comment-15334041 ]
Paulo Motta commented on CASSANDRA-12015: ----------------------------------------- while picking replicas from the same DC/rack is definitely useful, I'm not sure sorting replicas by dynamic snitch within the same rack/dc will buy us many benefits here for bulk operation like streaming. A simple fix here would be to use the current AbstractEndpointSnitch.sortByProximity instead, that will only sort replicas by rack/dc, which should pick primary replicas for each range and that should already yield a reasonable load distribution. > Rebuilding from another DC should use different sources > ------------------------------------------------------- > > Key: CASSANDRA-12015 > URL: https://issues.apache.org/jira/browse/CASSANDRA-12015 > Project: Cassandra > Issue Type: Improvement > Reporter: Fabien Rousseau > > Currently, when adding a new DC (ex: DC2) and rebuilding it from an existing > DC (ex: DC1), only the closest replica is used as a "source of data". > It works but is not optimal, because in case of an RF=3 and 3 nodes cluster, > only one node in DC1 is streaming the data to DC2. > To build the new DC in a reasonable time, it would be better, in that case, > to stream from multiple sources, thus distributing more evenly the load. -- This message was sent by Atlassian JIRA (v6.3.4#6332)