1 |
On Saturday 10 March 2007, mwq <mwq@××.pl> wrote about '[gentoo-user] |
2 |
RAID': |
3 |
> Imagine such a situation: I have two |
4 |
> hard drives but drive A is twice faster when reading and writing then |
5 |
> drive B. I want to make RAID 0 using A and B. Why are the stripes sizes |
6 |
> on both drives excacly the same? |
7 |
|
8 |
If stripe sizes aren't consistent, it's takes hundreds more CPU cycles (and |
9 |
a separate code path) to determine where to write a block. This |
10 |
calculation has to be repeated every time a block is written so, it can |
11 |
easily end up limiting your total throughput by making an I/O operation be |
12 |
CPU bound. |
13 |
|
14 |
> I think that using twice |
15 |
> greater stripe on A gives more speed then using equal stripes. |
16 |
|
17 |
If A is also twice as big, you can do as Dan Farrell suggested to get some |
18 |
increased performance. If A is the same size, you'd be sacrificing speed |
19 |
across the whole raid-ed device to potentially gain some speed in the |
20 |
sections where you can make A's segments bigger. |
21 |
|
22 |
-- |
23 |
Boyd Stephen Smith Jr. ,= ,-_-. =. |
24 |
bss03@××××××××××.net ((_/)o o(\_)) |
25 |
ICQ: 514984 YM/AIM: DaTwinkDaddy `-'(. .)`-' |
26 |
http://iguanasuicide.org/ \_/ |
27 |
New GPG Key! Old key expires 2007-03-25. Upgrade NOW! |