[
https://issues.apache.org/jira/browse/HDDS-15393?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hui Fei updated HDDS-15393:
---------------------------
Target Version/s: 2.3.0 (was: 2.2.0)
> Implement DAG Diff with multi-stage sequential read
> ---------------------------------------------------
>
> Key: HDDS-15393
> URL: https://issues.apache.org/jira/browse/HDDS-15393
> Project: Apache Ozone
> Issue Type: Sub-task
> Reporter: Saketa Chalamchala
> Assignee: Saketa Chalamchala
> Priority: Major
>
> Implement a multi-stage scan for an optimized DAG based diff:
> 1. Sequential scan of `toSnapshot` diff SSTs via K-way merge reads key
> entries into a `newList` and track the diff candidates.
> 2. Full table scan of `toSnapshot.directoryTable` to build intermediate
> column families used for `toSnapshot` key's path resolution.
> 3. Full table scan of `fromSnapshot.directoryTable` to build intermediate
> column families used for `fromSnapshot` key's path resolution and `oldList`
> population.
> 4. Batch point lookups (`multiGet`) of `fromSnapshot.fileTable` for diff
> candidates and read then into `oldList`.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
