Shilun Fan created HDDS-15630:
---------------------------------
Summary: [DiskBalancer] Execute batch DiskBalancer CLI requests
concurrently
Key: HDDS-15630
URL: https://issues.apache.org/jira/browse/HDDS-15630
Project: Apache Ozone
Issue Type: Sub-task
Components: Ozone Datanode
Reporter: Shilun Fan
Assignee: Shilun Fan
DiskBalancer CLI supports batch operations against multiple datanodes, for
example when users provide multiple datanode addresses or use
--in-service-datanodes.
Currently, these batch requests are executed sequentially. For large clusters,
commands such as start, stop, update, status, or report can take a long time
because the CLI waits for one datanode request to finish before sending the
next one.
This improvement proposes to execute batch DiskBalancer CLI requests
concurrently with a bounded thread pool. The CLI should preserve the existing
user-facing behavior as much as possible, including clear per-datanode
success/failure output and deterministic output ordering where applicable.
The goal is to reduce the latency of batch DiskBalancer operations while
keeping the implementation safe and predictable.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
