How to identify disk performance bottlenecks?

[SAKET_5]

Hi All,
Server Specs:
Proliant ML530
Pentium III 930Mhz
654MB RAM
Smart Array 4200 Controller (4 channels)
Windows 2000 Server with SP4 and post SP4 patches/hotfixes
SYSTEM DISK (C:\)- RAID 1 (2*9GB Ultra2 SCSI)
3 Data partitions each a RAID 5 set with 4*36.4GB Ultra 2 SCSI drives spread over 4 channels.
We suspect performance problems with the drives. Backup of the data partitions have started taking a long time. Windows 2000 Performance monitor displays the averagde disk read queue length counter in 90-100 range for most of the time for one of the data partitions that take long time for backup. What specifically does this mean if the queue length remains quite high for a logical volume? We have confirmed that the network setup, CPU utilisation and memory utilisation patterns are not odd.
How do we completely isolate the problems with drives?
Anyone, any help would be greatly appreciated.
Regards,

[Marino Meloni_1]

hi,

read these documents, it may be usefull

http://h20015.www2.hp.com/en/document.jhtml?lc=en&docName=lpg50460

http://h20015.www2.hp.com/en/document.jhtml?lc=en&docName=lpg50032

marino