I tried measuring virtual disk performance using bonnie++.
Case 1 result :
Using uid:0, gid:0.
Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Version 1.03e ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
TEST_IO 1G 91486 10 31793 2 71461 2 166.0 1
TEST_IO,1G,,,91486,10,31793,2,,,71461,2,166.0,1,,,,,,,,,,,,,
Case 2 Result :
Using uid:0, gid:0.
Writing intelligently...done
Rewriting...done
Reading intelligently...done
start 'em...done...done...done...
Version 1.03e ------Sequential Output------ --Sequential Input- --Random-
-Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
TEST_DP_IO 1G 104599 22 40353 6 99579 6 163.6 0
TEST_DP_IO,1G,,,104599,22,40353,6,,,99579,6,163.6,0,,,,,,,,,,,,,
In case 2 , I get better performance (correct me If I am wrong).
So I interpret results in this way:
Case 1 :
91486 = 89 MBps
31793 = 31.04 MBps
71461 = 69.7 MBps
166 IOPS
Case 2:
104599 = 102 MBps Write
40353 = 39.4 MBps Re-Write
99579 = 97.2 Read
163.6 IOPS
As you can see, write, Re-Write, Read performance is better in Case 2 except IOPS?? Should IOPs numbers must be greater than Case 1 ?? Is it valid to have lesser IOPS and more performance in Write, Read. ??
Thx