zeroshade commented on issue #37:
URL: https://github.com/apache/arrow-go/issues/37#issuecomment-5501445110

   Closing this as stale — I asked on 2026-07-11 whether the slowness still 
reproduced after the many reader performance improvements since 2024, and we 
didn't hear back.
   
   For anyone landing here with the same question: pass an `io.ReaderAt` that 
does S3 ranged GETs (for example https://github.com/wolfeidau/s3iofs) to 
`file.NewParquetReader`. Dropping to `file.NewPageReader` yourself is not worth 
it. If throughput is poor, the things worth checking are 
`ReaderProperties.BufferedStreamEnabled`, reading only the row groups and 
columns you need, and whether the client is doing many small ranged reads 
rather than a few large ones.
   
   @Zeeyi13 if 12 minutes for a 140MB file is still what you see on a current 
release, please open a new issue with the reader setup and I'll dig into where 
the time is going — that number is far worse than expected and I'd like to 
understand it.


-- 
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]

Reply via email to