drbd ver : version: 8.2.6 (api:88/proto:86-88)<br><br>Tests performed: <br>ipref shows 125MB/s~ , pureftpd also shows 125MB/s~<br><br>physical -> drbd : full 4GB resync = 105MB/s~ which also equals to , physical -> drbd -> ext3 , in cs=standalone/WFconnection mode = 105MB/s~<br>
<br>standalone/WFconnection test was done using, dd and bonnie++ , bonnie++ shows about 10MB/s less write performence (more rigorous test):<br>------------------------------------------------------------------------------------------------------------------<br>
time dd if=/dev/zero of=./testfile bs=16384 count=500000 <br>500000+0 records in<br>500000+0 records out<br>8192000000 bytes (8.2 GB) copied, 78.5591 seconds, 104 MB/s<br><br>real 1m18.971s<br>user 0m0.376s<br>sys 0m32.726s<br>
<br>bonnie++ -u 0 -n 0 -s 7180 -f -b -d ./<br>Using uid:0, gid:0.<br>Writing intelligently...done<br>Rewriting...done<br>Reading intelligently...done<br>start 'em...done...done...done...<br>Version 1.03 ------Sequential Output------ --Sequential Input- --Random-<br>
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--<br>Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP<br>cluster2.loca 7180M 89458 46 61011 29 157652 15 658.3 0<br>
cluster2.local,7180M,,,89458,46,61011,29,,,157652,15,658.3,0,,,,,,,,,,,,,<br><br>89MB/s~ write, 157MB/s~ read<br><br>------------------------------------------------------------------------------------------------------------------<br>
***** Now the bottleneck is when in **** primary/primary , or primary/secondary *** :<br>-------------------------------------------------------------------------------------------------------------------<br><br>time dd if=/dev/zero of=./testfile bs=16384 count=500000 <br>
500000+0 records in<br>500000+0 records out<br>8192000000 bytes (8.2 GB) copied, 100.704 seconds, 81.3 MB/s<br><br>bonnie++ -u 0 -n 0 -s 7180 -f -b -d ./<br><br>Using uid:0, gid:0.<br>Writing intelligently...done<br>Rewriting...done<br>
Reading intelligently...done<br>start 'em...done...done...done...<br>Version 1.03 ------Sequential Output------ --Sequential Input- --Random-<br> -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--<br>
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP<br>cluster1.loca 7180M 54283 17 59925 20 158998 15 583.0 0<br>cluster1.local,7180M,,,54283,17,59925,20,,,158998,15,583.0,0,,,,,,,,,,,,,<br>
<br>55MB/s~ write / 159MB/s~ read<br>-----------------------------------------------------------------------------------------------------------------------------------------<br>why the 30-40MB/s difference , compared to resync or standalone/WFconnection mode speed? <br>
<br>what operations in normal I/O activity can affect performance VS drbd resync operation? and how can i fix them ?<br>if resync is done via the network and it operates at speeds equal to standalone mode , what could hinder performance in normal primary/secondary , primary/primary mode like this?<br>
<br>btw - I have no-md-flushes and no-disk-flushes options on, since without that i am lucky to even get more than 10MB/s write speed , but you probably know about that...<br><br>All the best , Marcelo.<br>