[
https://issues.apache.org/jira/browse/LUCENE-9535?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17200946#comment-17200946
]
Adrien Grand commented on LUCENE-9535:
--------------------------------------
I might have found something. When profiling indexing I noticed some contention
in {{DocumentsWriterFlushControl#doAfterDocument}}, which happens to
transitively call {{IndexingChain#ramBytesUsed}}, which was changed in
LUCENE-9511 to call {{StoredFieldsWriter#ramBytesUsed}}. And
{{StoredFieldsWriter#ramBytesUsed}} calls
{{ByteBuffersDataOutput#ramBytesUsed}} which is a bit slow since it iterates
over all pages. So we might have increased contention on
{{DocumentsWriterFlushControl#doAfterDocument}} in LUCENE-9511, and this is
only noticeable on Mike's beast because of the very high number of indexing
threads (36). I opened https://github.com/apache/lucene-solr/pull/1917.
> Investigate recent indexing slowdown for wikimedium documents
> -------------------------------------------------------------
>
> Key: LUCENE-9535
> URL: https://issues.apache.org/jira/browse/LUCENE-9535
> Project: Lucene - Core
> Issue Type: Task
> Reporter: Adrien Grand
> Priority: Minor
> Attachments: cpu_profile.svg
>
> Time Spent: 20m
> Remaining Estimate: 0h
>
> Nightly benchmarks report a ~10% slowdown for 1kB documents as of September
> 9th: [http://people.apache.org/~mikemccand/lucenebench/indexing.html].
> On that day, we added stored fields in DWPT accounting (LUCENE-9511), so I
> first thought this could be due to smaller flushed segments and more merging,
> but I still wonder whether there's something else. The benchmark runs with
> 8GB of heap, 2GB of RAM buffer and 36 indexing threads. So it's about 2GB/36
> = 57MB of RAM buffer per thread in the worst-case scenario that all DWPTs get
> full at the same time. Stored fields account for about 0.7MB of memory, or 1%
> of the indexing buffer size. How can a 1% reduction of buffering capacity
> explain a 10% indexing slowdown? I looked into this further by running
> indexing benchmarks locally with 8 indexing threads and 128MB of indexing
> buffer memory, which would make this issue even more apparent if the smaller
> RAM buffer was the cause, but I'm not seeing a regression and actually I'm
> seeing similar number of flushes when I disabled memory accounting for stored
> fields.
> I ran indexing under a profiler to see whether something else could cause
> this slowdown, e.g. slow implementations of ramBytesUsed on stored fields
> writers, but nothing surprising showed up and the profile looked just like I
> would have expected.
> Another question I have is why the 4kB benchmark is not affected at all.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
