hunterpayne commented on issue #40083:
URL: https://github.com/apache/arrow/issues/40083#issuecomment-1944480365

   That sounds like a long time from now.  This is this correct?  Also from the 
POV of Java code I have to ask why this is the case.  You can get access to 
memcpy (the syscall you really want) without all of this complexity.  
System.arrayCopy is just a bounds checked memcpy.  Seems to me that you are 
making this overly complex due to the libraries you are using.
   
   PS I worked on database codebases for about a decade.  Some of that time was 
in the DB kernel.  I do know of what I speak.
   PPS If your buffers aren't a multiple of the size of a virtual page in the 
OS (usually 4k), then accessing memcpy doesn't really help performance.  Other 
factors will overwhelm any minimal performance improvements you get by 
bypassing the Java heap.
   


-- 
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