pitrou commented on issue #41604:
URL: https://github.com/apache/arrow/issues/41604#issuecomment-2112782351

   > But the caller chooses `maxBytes` :)
   
   Right, but they have no idea what value could be reasonable. The caller 
could think "ok, I know that some filesystems really like large chunk sizes, so 
I'm going to allocate 100 MiB", only to get 64 kiB as result.
   
   Or conversely, the caller could think "ok, let's be conservative so as not 
to waste memory unduly, so I'm going to allocate 512 kiB", and S3 will perform 
awfully.
   
   The point here is to let the file implementation inform the caller before 
making any impacting decision (such as allocating a large memory area).
   
   > Which one you think is simpler?
   
   Both are simple conceptually and should be simple implementation-wise. They 
have different implications though.
   


-- 
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