Email Subscription Notifications Suspended Temporarily
We are in the process of making navigation in the Servers and Operating Systems forums simpler and more direct. While doing this, we have to temporarily suspend email notifications for subscriptions. If you are subscribed to one or more discussion boards or blogs in the community, please check them daily to see new content. Notifications will be turned back on in a few days. We apologize for any inconvenience this may cause. Thanks, Warren_Admin
Disk Enclosures
cancel
Showing results for 
Search instead for 
Did you mean: 

raid1 vg I/O speed

Paolo Gilli
Frequent Advisor

raid1 vg I/O speed

hi,
I'am working on an HP9000 4000/n, with 6 SC10 enclosure and 1 FC60 controller.
I need create a raid1 VG with 2 18Gb hd.
In this vg I will create a lv for temporary tablespace for oracle 8i.
Is there different I/O speed using PE_size=32 instead PE_size=4 (default)?

which is the best PE_SIZE?
Thank
2 REPLIES
Sanjay Kumar Suri
Honored Contributor

Re: raid1 vg I/O speed

The one-to-one mapping of physical to logical volumes is necessary to access the physical location and to perform I/O operations.

This value is normally set to 4 megabytes. This allows for a maximum size of 256 gigabytes for physical and logical volumes.

The physical extent size should only be increased if you need larger logical volumes than 256 gigabytes.

sks
A rigid mind is very sure, but often wrong. A flexible mind is generally unsure, but often right.
A. Clay Stephenson
Acclaimed Contributor

Re: raid1 vg I/O speed

With one possible (and very dubious exception) the answer is no. PE size has no impact upon performance; it should be chosen to allow the largest anticatped physical volume in a given volume group. The exception is if you are using extent-based striping. In this case, the smallest possible PE size should be used in order to distribute the IO more quickly but even here the smallest possible PE size (1MB) is much too large to be a good io stripe size.
If it ain't broke, I can fix that.