Hi All

I have two similar database (Oracle 9)and application server (weblogic 8.1)configured with same setup in my Red Hat 5.2 Enterprise edition Servers and I'm running a dd command ,to be more specific
the two servers performance is of this type.


time dd if=/dev/zero of=./test.zero bs=1024 count=1000000

real 0m12.92s real 0m3.22s
user 0m0.41s user 0m1.17s
sys 0m11.74s sys 0m2.81s



This is what I see those are the two comparisons side by side like real of one server vs the real of server 2 , Please bear with the space . The application and database is stable I checked but the performance of those two boxes is like this , Can anybody tell me what is a permanent solution like modify any Unix settings because reboot fixed it once and reboot created it the other time and I dont wan't to repeat reboot steps. Thanks guys.

There are a number of questions that need to be asked.

1) Are the servers the same (Same CPU, RAM, etc......)
2) Were the application server and database shutdown whilst you performed the tests, if not they could impact the test.
3) You seem to be looking at disk performance, are the servers using the same disk setup.

Looking at the figures you quoted the first set seems to come from the Oracle database server the second set from the application server. Assuming you didn't shutdown the database and application server then I would expect the task to take longer on the db server because the server is having to perform IO for the database and the test your performing.

If the servers use different disks then that would also account for the performance difference potentially.