1 |
On Mon, Dec 1, 2008 at 5:07 PM, Anthony Metcalf <nevyn@××××××××××.uk> wrote: |
2 |
> Ignoring the MS/*nix argument for a moment..... |
3 |
> |
4 |
> How does fragmentation work on hardware RAID? |
5 |
> |
6 |
> The normal argument of more fragmentation = more seeks = worse |
7 |
> performance would seem to be eliminated by a controller with a large |
8 |
> cache and lots of drives (say a Dell/Equallogic PS5000E in RAID 10.....). |
9 |
> |
10 |
> A |
11 |
|
12 |
Assuming a level that performs striping, it would be alleviated, yes, |
13 |
but even striping is at its fastest when performing sequential reads, |
14 |
it just reads sequentially from more than one disk at a time... but |
15 |
the fact that one disk will likely be reading while another performs a |
16 |
seek would probably reduce the impact of the seeks considerably. |
17 |
|
18 |
(all of this is, of course, purely pulling from my understanding of |
19 |
raid and how it operates and could very well be 100% incorrect) |
20 |
|
21 |
-- |
22 |
Poison [BLX] |
23 |
Joshua M. Murphy |