abhishekagarwal87 opened a new pull request #12269:
URL: https://github.com/apache/druid/pull/12269


   Using `InputFormat` instead of `ParseSpec` can be very slow if complex 
flattening scheme is used. This is because with InputFormat, we create an 
`InputEntityReader` for every stream chunk. It was assumed that creating the 
reader is a cheap operation but it is not in some cases. 
   
   Here are two flame graphs to illustrate the difference
   Task runner thread with InputFormat
   ![Screenshot 2022-02-18 at 1 42 58 
PM](https://user-images.githubusercontent.com/1477457/154686173-9f8551b0-54b0-4986-b601-ae54ed62684d.png)
   
   Task runner thread with ParseSpec
   ![Screenshot 2022-02-18 at 1 41 46 
PM](https://user-images.githubusercontent.com/1477457/154686201-03b5c9eb-efaa-4f16-aefa-74530da6dcf7.png)
   
   Instead of fixing the JsonParser, I have added a new class called 
`SettableByteEntity` that allows one to add new binary data without creating a 
new InputStream. So we can create an InputEntityReader just once instead of 
creating it for every stream chunk. 
   
   This PR has:
   - [ ] been self-reviewed.
      - [ ] using the [concurrency 
checklist](https://github.com/apache/druid/blob/master/dev/code-review/concurrency.md)
 (Remove this item if the PR doesn't have any relation to concurrency.)
   - [ ] added documentation for new or modified features or behaviors.
   - [ ] added Javadocs for most classes and all non-trivial methods. Linked 
related entities via Javadoc links.
   - [ ] added or updated version, license, or notice information in 
[licenses.yaml](https://github.com/apache/druid/blob/master/dev/license.md)
   - [ ] added comments explaining the "why" and the intent of the code 
wherever would not be obvious for an unfamiliar reader.
   - [ ] added unit tests or modified existing tests to cover new code paths, 
ensuring the threshold for [code 
coverage](https://github.com/apache/druid/blob/master/dev/code-review/code-coverage.md)
 is met.
   - [ ] added integration tests.
   - [ ] been tested in a test Druid cluster.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]



---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to