LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Solaris / OpenSolaris (https://www.linuxquestions.org/questions/solaris-opensolaris-20/)
-   -   Disk performance - is that true ? (https://www.linuxquestions.org/questions/solaris-opensolaris-20/disk-performance-is-that-true-706121/)

czezz 02-20-2009 02:26 AM

Disk performance - is that true ?
 
Can anyone explain me is this true: that disk the smaller is the faster is ?

What my situation is:
I have 2 types of disks. Same vemdor; same firware revision; same RPM speed; same model but size diffrent (146 G and 300 G)

Code:

disk 1 = HITACHI HUS1514FBSUN146G 2A02
disk 2 = HITACHI HUS1530FBSUN300G 2A02

*Group of disks 1 are going to be joined into a RAID 1
*Group of disks 2 are going to be joined into a RAID 1
*all of them are going to be plugged in to a Sun StorageTek 3510.

jlliagre 02-20-2009 04:54 AM

It might be true. What state the manufacturer specs ?

Randux 02-20-2009 05:48 AM

Not really. The faster it spins the faster it is in general though.

salasi 02-20-2009 06:43 AM

Quote:

Originally Posted by czezz (Post 3450715)
Can anyone explain me is this true: that disk the smaller is the faster is ?

By smaller do you mean number of Gigabytes? (And not physical size)

Within the same series (same phys size), different sizes of disks are normally made by having different numbers of platter surfaces active, so a 300 G drive may have two and a 450 three.

The speed (how measured? just don't worry about that yet) decreases as you get closer to the centre of the disk, so the outer sectors are faster.

However, sometimes, largely for marketing reasons, but sometimes to deal with batches of platters which have defective regions, the manufacturers sometimes sell drives with lower-than-their-potential-capacity. When they do this, they will preferentially disable the slower parts of the disk. So if, in the previous example, if you saw a 400 G disk they would have disabled some, probably the slower parts, of a 450 G disk (and probably that 450 G disk, using 3 surfaces, was probably physically a 600 G/4 surface disk with a whole bad surface disabled or maybe even without magnetic coating on that one surface).

The other way of looking at this phenomenon is that a larger percentage of the disk is the fast stuff.


All times are GMT -5. The time now is 09:54 AM.