[
https://issues.apache.org/jira/browse/AVRO-557?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12897024#action_12897024
]
Scott Carey commented on AVRO-557:
----------------------------------
The one that fixed it for me is not your second one, but the first one you
posted.
Don't trust the current performance numbers from the test. I'll upload a
partial patch that only includes enhancements to Perf.java with better
measurements right away.
For me, using your _first_ patch that only caches the resolver, I get:
GenericReaderTest: 4413 ms, 0.9439455178163628 million entries/sec.
36.63489596549099 million bytes/sec
GenericReaderOneTimeUseReaderTest: 4563 ms, 0.9130533872713805 million
entries/sec. 35.44198664989503 million bytes/sec
GenericReaderOneTimeUseDecoderTest: 4476 ms, 0.9307708856028631 million
entries/sec. 36.128051198318644 million bytes/sec
GenericReaderOneTimeUseTest: 4644 ms, 0.8970147804525915 million entries/sec.
34.812096233011566 million bytes/sec
> Speed up one-time data decoding
> -------------------------------
>
> Key: AVRO-557
> URL: https://issues.apache.org/jira/browse/AVRO-557
> Project: Avro
> Issue Type: Improvement
> Components: java
> Affects Versions: 1.3.2
> Reporter: Kevin Oliver
> Assignee: Doug Cutting
> Fix For: 1.4.0
>
> Attachments: AVRO-557.patch, AVRO-557.patch, AVRO-557.patch,
> AVRO-557.patch
>
>
> There are big gains to be had in performance when using a BinaryDecoder and a
> GenericDatumReader just one time. This is due to the relatively expensive
> parsing and initialization that came with 1.3. Patch with example code and a
> Perf harness to follow.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
