返回
-
标题
Dell PERC Controller Cache Settings for improved disk performance -
说明
Slow disk performance of the repository when the repository is located on local storage or DAS (Direct Attach Storage) connected to a Dell PERC disk controller.
How to identify this problem:
- Disk activity is at 100% but the read speed is very slow.
- In Task Manager | Resources Monitor | Disk| Storage shows 100% disk activity on the repository drive but the disk I/O is very low, only a few MB/s.
- SQL attachability taking hours to complete
- Repository mounting job taking a very long time
-
原因
The Controller Cache is Disabled or not set to the optimal performance for a Rapid Recovery Repository. You can check the settings from OMSA (Dell OpenManage Server Administrator) | Storage | Virtual Disks. The Read Policy is set to: No Read Ahead and the Write Policy is set to: Write Through.
This is not the optimal settings for a Rapid Recovery Repository. NOTE: The Disk Cache Policy should be Disabled if your Controller has it's own cache. This references the disks own cache.
-
解决办法
- Set the Read Policy: Adaptive Read Ahead Or Read Ahead starting since PERC 8 and 9.
NOTE: Previous generations of PERC controllers support read policy settings of No Read Ahead , Read Ahead, and Adaptive Read Ahead. With PERC 8 and PERC 9, the Read Ahead and Adaptive Read Ahead settings are functionally equivalent at the controller level. For backward compatibility purposes, some systems management interfaces and PERC 8 and 9 controllers still allow setting the read policy to Adaptive Read Ahead. While it is possible to set Read Ahead or Adaptive Read Ahead on PERC 8 or PERC 9, there is no functional difference.
- Set the Write Policy: Write Back
- Disk Cache Policy: Disabled
- Adaptive Read Ahead—Begins using Read-Ahead if the two most recent disk accesses occurred in sequential sectors. If the read requests are random, the controller reverts to No Read Ahead mode. If the following reads are sequential the cache will go back to using Read Ahead.
- In Write-Back caching, the controller sends a data transfer completion signal to the host when the controller cache has received all the data in a transaction. The controller then writes the cached data to the storage device in the background.
- The risk of using Write-Back cache is that the cached data can be lost if there is a power failure before it is written to the storage device. This risk is mitigated by using a Non-Volatile Cache on the PERC H710, H710P or H810 card.
- If you need any assistance with the cache settings, please call the Quest support or open the ticket online.
Is your Rapid Recovery solution working as efficiently as possible? Learn more about our Rapid Recovery Health Check.