Symptom:
We have identify an I/O performance downgrade after we install and
consolidate all the Transactional DBs of the company.
Using perfmon, the values of % Disk Time (Physical & Logical) on the drive
that has all de physical BDs report an average use arround 2000% (yes two
thousand), in other had, the avg values of avg disk qeue length (read &
write) are over expected values, reporting on average .150 & .300 ms.
We are conducting a report on FC & SAN use.
Question:
- What can we do to understand what is happening with the server and improve
the performance?
The best implementation is to isolate the data disk and the log disks, I
mean, put it on differente arrays on the SAN...
When you say that you are conducting a report , it means Reporting Services???