1 |
I have used the 3Ware (now AMCC) products with considerable success. They |
2 |
can be confiuged to raid 1, raid 10, and raid 5 formats as well as a bod |
3 |
configuration. |
4 |
|
5 |
|
6 |
On Thu, 2 Mar 2006, Jeroen Geilman wrote: |
7 |
|
8 |
> Patrick Lauer wrote: |
9 |
> > On Wed, 2006-03-01 at 13:40 +0100, Marton Gabor wrote: |
10 |
> > |
11 |
> >> Hi! |
12 |
> >> |
13 |
> >> I'm going to recieve 4x250Gb SATA disks to our new server, and my first |
14 |
> >> idea was to make 2xRAID1 and then make 1xRAID0 out of the RAID1 arrays |
15 |
> >> using Linux software raid so that I have our data mirrored and still I |
16 |
> >> can use 500Gb storage space and handle it as one big "disk". |
17 |
> >> So my questions would be: |
18 |
> >> - could someone give me a good howto? Sorry, but I have never had |
19 |
> >> the chance to make a RAID array before and I have no experience and |
20 |
> >> Google doesn't seem to be helpful in this case. |
21 |
> >> - do I need to make a /boot partition which is not part of any |
22 |
> >> arrays or will grub boot from raid1+0? |
23 |
> >> |
24 |
> > grub can boot from raid1, raid0+1 will need a (small) boot partition. |
25 |
> > |
26 |
> > With 4 disks you could also build a raid5 with little overhead, takes a |
27 |
> > tad more cpu and gives you 750G capacity (or 500 with one hotspare) |
28 |
> > |
29 |
> Take note that both using RAID5 and RAID10 in software will use a |
30 |
> significant amount of CPU*; normally speaking (in a hardware |
31 |
> configuration) RAID10 would outperform RAID5 by 30% or more, but since |
32 |
> it's in software the RAID0 has to be layered on top of the RAID1, |
33 |
> increasing its overhead by no small amount. |
34 |
> |
35 |
> I'd go with the RAID1 with LVM solution mentioned earlier if you intend |
36 |
> to retain any performance worth mentioning. |
37 |
> |
38 |
> If there are decent Linux drivers for it, I'd highly recommend a RAID |
39 |
> card that can do RAID5 or RAID10 in hardware. |
40 |
> |
41 |
> *Actually, the RAID10 solution won't use nearly as much CPU as the |
42 |
> RAID5, but the RAID10 will spend a lot more time waiting on disk I/O, so |
43 |
> the net result will likely be similar, if not actually worse. |
44 |
> |
45 |
> |
46 |
> |
47 |
> |
48 |
|
49 |
-- |
50 |
|
51 |
-- |
52 |
gentoo-server@g.o mailing list |