Jackie-Jiang commented on pull request #7076:
URL: https://github.com/apache/incubator-pinot/pull/7076#issuecomment-872631434


   @tdunning Thanks for the explanation.
   I understand the error is unavoidable, but IMO the test is legit and just 
mimics the behavior of utilizing t-digest to pre-grouping the numbers and get 
the compressed statistics instead of storing all the numbers. From user's 
perspective, grouping the numbers in different order should give very close 
result.
   The concern here is that with exactly the same way of grouping numbers, we 
observed larger error rate with the new version compared with the old version. 
We want to justify this behavior. Is this caused by better/more strict 
compression in the new version? What is the benefit of upgrading to the new 
version?


-- 
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