windowswideopen.com

Here are a list of some of the potential causes of poor SAN performance:

1. HBA firmware, drivers, configuration settings
2. Fibre Channel switch configuration issues
3. Contention in the SAN between the server and storage
4. Settings on the storage unit
5. “Split I/O” on hardware RAID volumes
6. Conflicting software or drivers

Troubleshooting steps to take:

1. Update outdated HBA firmware and drivers as they may resolve the issue.
2. Contact the relevant HBA card vendor to help detect non-standard HBA drive settings.
3. Review HBA BIOS settings
4. Check queue depth settings for Qlogic and Emulex HBAs
5. Check all points between the server and the storage unit. If a fibre channel switch is involved then check port logs on the switch for CRC, encoding (ENC), timeout, loss of light, and other errors.
6. If the storage unit has a reporting facility then check all logs for errors. Also check the cache settings such as write back” vs. “write through” as cache settings can have a huge impact on performance.
7. Run a performance log that includes the physical disk, processor, and system objects. Look for high “disk queue length”. Note that the disk queue length for a given SAN volume is the total number divided by number of disk spindles per volume. A sustained reading over 2 for “disk queue length” indicates congestion.

Also see:

Troubleshoot Windows Disk Problems
Troubleshooting Windows SAN Problems

Disks Fibre Channel SAN PerformanceDisks Fibre Channel SAN Performance

Filed under: Disks