[
https://issues.apache.org/jira/browse/LUCENE-9378?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17198432#comment-17198432
]
Adrien Grand commented on LUCENE-9378:
--------------------------------------
I'm surprised it's slower since other cases that were about scanning the entire
data on non-compressible data reported speedups with this new format. I wonder
why this one is different.
Indeed by asking for use-cases, I was trying to understand whether the solution
to this would be to introduce a new type of doc values rather than disabling
compression on binary doc values, or even a new format like we're considering
for feature vectors.
I'm not sure what the path forward is. This trade-off we're making with binary
doc values is not unique, it's the same one we're making with postings where
you can't decode a single doc ID, you have to decode whole blocks of 128
values. While reverting to the previous format would address the performance
regressions that have been reported here, I'm a bit unhappy of abandoning
completely the idea of leveraging redundancy we might see across values. Maybe
we should look into using dictionaries like we do for stored fields in order to
be able to achieve good compression ratios with smaller block sizes.
> Configurable compression for BinaryDocValues
> --------------------------------------------
>
> Key: LUCENE-9378
> URL: https://issues.apache.org/jira/browse/LUCENE-9378
> Project: Lucene - Core
> Issue Type: Improvement
> Reporter: Viral Gandhi
> Priority: Minor
> Attachments: hotspots-v76x.png, hotspots-v76x.png, hotspots-v76x.png,
> hotspots-v76x.png, hotspots-v76x.png, hotspots-v77x.png, hotspots-v77x.png,
> hotspots-v77x.png, hotspots-v77x.png, image-2020-06-12-22-17-30-339.png,
> image-2020-06-12-22-17-53-961.png, image-2020-06-12-22-18-24-527.png,
> image-2020-06-12-22-18-48-919.png, snapshot-v77x.nps, snapshot-v77x.nps,
> snapshot-v77x.nps, snapshots-v76x.nps, snapshots-v76x.nps, snapshots-v76x.nps
>
> Time Spent: 4h 10m
> Remaining Estimate: 0h
>
> Lucene 8.5.1 includes a change to always [compress
> BinaryDocValues|https://issues.apache.org/jira/browse/LUCENE-9211]. This
> caused (~30%) reduction in our red-line QPS (throughput).
> We think users should be given some way to opt-in for this compression
> feature instead of always being enabled which can have a substantial query
> time cost as we saw during our upgrade. [~mikemccand] suggested one possible
> approach by introducing a *mode* in Lucene80DocValuesFormat (COMPRESSED and
> UNCOMPRESSED) and allowing users to create a custom Codec subclassing the
> default Codec and pick the format they want.
> Idea is similar to Lucene50StoredFieldsFormat which has two modes,
> Mode.BEST_SPEED and Mode.BEST_COMPRESSION.
> Here's related issues for adding benchmark covering BINARY doc values
> query-time performance - [https://github.com/mikemccand/luceneutil/issues/61]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@lucene.apache.org
For additional commands, e-mail: issues-h...@lucene.apache.org
