You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@cassandra.apache.org by "Johnny Miller (Jira)" <ji...@apache.org> on 2022/12/08 10:30:00 UTC

[jira] [Created] (CASSANDRA-18103) Allow different compaction strategies per-DC

Johnny Miller created CASSANDRA-18103:
-----------------------------------------

             Summary: Allow different compaction strategies per-DC
                 Key: CASSANDRA-18103
                 URL: https://issues.apache.org/jira/browse/CASSANDRA-18103
             Project: Cassandra
          Issue Type: Improvement
            Reporter: Johnny Miller


I have a requirement for deploying an additional DC. The cluster is split between multiple DCS - on-prem and a cloud.

Several tables use LCS and perform fine on bare metal and not so well on the infrastructure allocated for the cloud DC.

The cloud deployment is intended to run offline analytical batch-type workloads where the criticality of read response times does not necessitate LCS. The cost of presenting appropriate storage for LCS is high and unnecessary for the system requirements or budget.

The JMX call to change compaction locally for testing/migrating compaction, unfortunately, does not survive restarts, schema changes etc.

It would be very helpful to indicate on the table what compaction strategy to use per dc or make the JMX change durable. 



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org