[jira] [Created] (HDDS-15271) Client should prioritize replicas with BCSID covering the blocks.

["Ivan Andika (Jira)" <ji...@apache.org>]

Ivan Andika created HDDS-15271:
----------------------------------

             Summary: Client should prioritize replicas with BCSID covering the blocks.
                 Key: HDDS-15271
                 URL: https://issues.apache.org/jira/browse/HDDS-15271
             Project: Apache Ozone
          Issue Type: Improvement
            Reporter: Ivan Andika
            Assignee: Ivan Andika


Currently, client read prioritizes based on the locality, datanode status (maintenance & decommission), etc. However, the client do not check whether the replica BCSID covers the block the client is trying to read. This causes BCSID_MISMATCH which triggers failover and increases read latency.

The idea of this patch is to also consider the BCSID as a hint (not a requirement) for client to pick a datanode. If a client requested a block with BCSID N, any datanodes that contains BCSID >= N should be prioritized over those that have datanodes BCSID < N. 

However, we need to note a few things
* We should not exclude the replicas with BCSID < N since the container replica BCSID might be stale (either container location cache is stale or container replica heartbeat has not been recorded by SCM). This means that although we will read from replicas with BCSID < N if the previous replicas with BCSID >= N have been replicated.
* We need to consider all BCSID >= N as being equal. So a replica 1 with BCSID N + 1 and replica 2 with BCSID N + 2 are the same even though replica 2 is more up-to-date. This should prevent hotspot.

We can include BCSID as the sorting requirements for client read.



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

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org