[
https://issues.apache.org/jira/browse/LUCENE-10502?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17536119#comment-17536119
]
ASF subversion and git services commented on LUCENE-10502:
----------------------------------------------------------
Commit bb56ce14e8b0a7f65d2d3dd4357443ea83118a0d in lucene's branch
refs/heads/branch_9x from Lu Xugang
[ https://gitbox.apache.org/repos/asf?p=lucene.git;h=bb56ce14e8b ]
LUCENE-10502: Use IndexedDISI to store docIds and DirectMonotonicWriter/Reader
to handle ordToDoc (#885)
Currently vector's all docs of all fields are fully loaded into memory (for
sparse cases).
This happens not only when we do vector search, but also when we open an index
to
load meta info for vector readers.
This patch instead uses IndexedDISI to store docIds and
DirectMonotonicWriter/Reader to
handle ordToDoc mapping. Benefits are reduced memory usage, and faster loading
of
meta info for vector readers.
Backport for #880
> Use IndexedDISI to store docIds and DirectMonotonicWriter/Reader to handle
> ordToDoc
> ------------------------------------------------------------------------------------
>
> Key: LUCENE-10502
> URL: https://issues.apache.org/jira/browse/LUCENE-10502
> Project: Lucene - Core
> Issue Type: Improvement
> Affects Versions: 9.1
> Reporter: Lu Xugang
> Priority: Major
> Time Spent: 15h 40m
> Remaining Estimate: 0h
>
> Since at search phase, vector's all docs of all fields will be fully loaded
> into memory, could we use IndexedDISI to store docIds and
> DirectMonotonicWriter/Reader to handle ordToDoc mapping?
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
