[
https://issues.apache.org/jira/browse/TIKA-2180?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15715310#comment-15715310
]
Tim Allison commented on TIKA-2180:
-----------------------------------
The file format (docx) seems to be the bottleneck.
When I ran 4 consumers against 4 input documents ("save as" from your initial
War and Peace .docx) on an 8 cpu machine, I got:
rtf 4.7 seconds
pdf 21 seconds
doc 6.7 seconds
txt 4.7 seconds
docx (legacy parser): I gave up -- machine pegged, 8gb memory used...thrashing
(FULL GC (Ergonomics))
docx (experimental SAX parser): 9 seconds (never got above 200MB)
When I ran it with one consumer thread:
docx (legacy parser): 67 seconds (6gb used)
docx (experimental SAX parser): 15 seconds (never got above 200MB)
> Multiple requests on Tika to extract text slows down
> ----------------------------------------------------
>
> Key: TIKA-2180
> URL: https://issues.apache.org/jira/browse/TIKA-2180
> Project: Tika
> Issue Type: Bug
> Components: server
> Affects Versions: 1.13, 1.14
> Environment: Windows OS, Open JDK, 4 core 32 GB RAM
> Reporter: Ashish Basran
>
> I observed that if I send multiple requests to Tika (eg.
> http://localhost:8080/tika) with around 5MB files, Tika is very slow in
> completing the action. I tried with ~20 random files, it took 170 seconds to
> process all the files in sequence. If I pass all files in parallel, it took
> around 780 seconds to process same set of files.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
