thinkharderdev opened a new issue, #5129:
URL: https://github.com/apache/arrow-datafusion/issues/5129

   **Is your feature request related to a problem or challenge? Please describe 
what you are trying to do.**
   A clear and concise description of what the problem is. Ex. I'm always 
frustrated when [...] 
   (This section helps Arrow developers understand the context and *why* for 
this feature, in addition to  the *what*)
   
   Current `FileStream` is a state machine that goes from Open (read metadata 
and build reader) -> Scan (scan decode `RecordBatch`) in sequence until all 
files are scanned. 
   
   For parquet in particular, the `Opener` may need to do multiple sequential 
IO requests:
   1. Read footer to get metadata start offset
   2. Read footer
   3. Fetch page/offset indexes
   
   Seems like it might be possible to parallelize the scan with opening the 
next file.
   
   **Describe the solution you'd like**
   A clear and concise description of what you want to happen.
   
   `FileStream::Scan` adds an `Option<FileOpenFuture>`. If polled in a scan 
state, try and pop another file and initialize it's `FileOpenFuture` (eg try 
and open the file in parallel) so metadata IO can be done in parallel to 
scanning of the current file. 
   
   **Describe alternatives you've considered**
   A clear and concise description of any alternative solutions or features 
you've considered.
   
   Not do anything
   
   **Additional context**
   Add any other context or screenshots about the feature request here.
   
   Not sure if this will be workable but want to experiment with 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