wy96f commented on issue #2845: ARTEMIS-2336 Use zero copy to replicate 
journal/page/large message file (AGAIN)
URL: https://github.com/apache/activemq-artemis/pull/2845#issuecomment-535807466
 
 
   @franz1981 
   > If both cases (with/without file region) are saturating the network, why 
the latter will take more time? The total amount of data sent should be the 
same...
   
   They're taking same time(~7 mins). As i said,
   ``
   In the case of -Dio.netty.file.region=true and master, log(something like 
this 2019-09-26 11:02:49,348 DEBUG 
[org.apache.activemq.artemis.core.replication.ReplicationManager] sending 
1048576 bytes on file xxxx) showed it took about 7 minutes to transfer files, 
then synchronization done message sent. However in the case of 
-Dio.netty.file.region=false, log showed it took about about 40 seconds to 
transfer files, then sync done message sent. 
   ``
   ``
   Yes, i saw the network was saturated(initial-replication-sync-timeout was 
set to 300000, otherwise replication failed due to timeout). Note that log 
showed 13:47:01,945 last page sent, and sar showed 01:47:37 PM queued up data 
still transferring and saturating network.
   ``
   Saturation lasted for ~7 mins.
   
   > Do you have tried master as well?
   
   The same. 
   
   

----------------------------------------------------------------
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

Reply via email to