[
https://issues.apache.org/jira/browse/HDDS-16179?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sreeja Chintalapati updated HDDS-16179:
---------------------------------------
Summary: Dry run command for container balancer with profiles (was: Add
logic to estimate balancer duration and iteration count)
> Dry run command for container balancer with profiles
> ----------------------------------------------------
>
> Key: HDDS-16179
> URL: https://issues.apache.org/jira/browse/HDDS-16179
> Project: Apache Ozone
> Issue Type: Sub-task
> Reporter: Sreeja Chintalapati
> Assignee: Sreeja Chintalapati
> Priority: Major
>
> h2. Problem
> Today, users can start the container balancer with {{{}ozone admin
> containerbalancer start{}}}, but there is no way to preview:
> * how much data needs to be moved
> * how long balancing may take
> * how different throttling settings affect throughput and duration
> This makes it hard to plan maintenance windows or choose safe balancer
> settings without trial and error.
> h2. Proposed Solution
> Add {{ozone admin containerbalancer dry-run}} that:
> # Fetches live datanode usage from SCM
> # Reads balancer limits and profile presets from the local Ozone
> configuration (including {{{}ozone-site.xml{}}})
> # Estimates, per profile:
> ** bytes to move
> ** per-iteration throughput
> ** estimated iterations
> ** upper-bound duration (with a +30% planning buffer)
> Introduce SLOW, MEDIUM, and FAST throttling profiles with new
> {{hdds.container.balancer.profile.*}} config keys for:
> * max datanode involvement per iteration
> * max bytes entering a target datanode
> * max bytes leaving a source datanode
> Add {{ContainerBalancerAdvisor}} to centralize cluster analysis, estimation,
> and validation logic used by dry-run.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
