tgravescs commented on issue #25299: [SPARK-27651][Core] Avoid the network when 
shuffle blocks are fetched from the same host
URL: https://github.com/apache/spark/pull/25299#issuecomment-558228346
 
 
   ok that makes sense. thanks.  So it seems we fetch all the local blocks in 
first and then the host local blocks and all that happens in the current thread 
while the remote block fetcher happens in separate.
   My concern could go both ways then. Now we are having to wait to fetch the 
local blocks first before the host local blocks where as before that could have 
been in parallel with the local blocks.  Or if we end up with a lot of host 
local blocks then within those fetches they happen serially now whereas they 
would have happened in parallel before.   
   Mostly just an observation to something we may want to enhance later if we 
see it as an issue.  

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to