1 |
I supported about 250 gentoo vm's using about 30 SAS 15K rpm 144GB |
2 |
drives awhile back. Drives were split into 14 disk RAID10 sets. Then |
3 |
each RAID10 set was split it into 200-500GB virtual drives, and the |
4 |
virtual machines were grouped into sets of 3-5 and matched with a |
5 |
virtual drive. Virtual machines on the same virtual drive were setup |
6 |
to use thin provisioning, so that only used up as much storage space |
7 |
as their data differed from the canonical gentoo os image which was |
8 |
usually less than 20%. The virtual drives were usually only 30-50% |
9 |
full and we could virtually provision 2TB+ of virtual machines on a |
10 |
single 500GB virtual drive. |
11 |
|
12 |
Don't underestimate what you can do with small drives, especially if |
13 |
they are fast and you have a lot of them.... |
14 |
|
15 |
On Thu, Jun 27, 2013 at 5:51 PM, Duncan <1i5t5.duncan@×××.net> wrote: |
16 |
> Mark Knecht posted on Sat, 22 Jun 2013 16:04:06 -0700 as excerpted: |
17 |
> |
18 |
>> Lastly, even if I completely buy into Duncan's well formed reasons about |
19 |
>> why RAID1 might be faster, using 500GB drives I see no single RAID |
20 |
>> solution for me other than RAID5/6. The real RAID1/RAID6 comparison from |
21 |
>> a storage standpoint would be a (conceptual) 3-drive RAID6 vs 3 drive |
22 |
>> RAID1. Both create 500GB of storage and can (conceptually) lose 2 drives |
23 |
>> and still recover data. However adding another drive to the RAID1 gains |
24 |
>> you more speed but no storage (buying into Duncan's points) vs adding |
25 |
>> storage to the RAID6 and probably reducing speed. As I need storage what |
26 |
>> other choices do I have? |
27 |
>> |
28 |
>> Answering myself, take the 5 drives, create two RAIDS - a 500GB |
29 |
>> 2-drive RAID1 for the system + VMs, and then a 3-drive RAID5 for video |
30 |
>> data maybe? I don't know... |
31 |
>> |
32 |
>> Or buy more hardware and do a 2 drive SSD RAID1 for the system, or |
33 |
>> a hardware RAID controller, etc. The options explode if I start buying |
34 |
>> more hardware. |
35 |
> |
36 |
> Finally getting back to this on what's my "weekend"... |
37 |
> |
38 |
> Unfortunately, given 900 gigs media data and 150 gigs of VMs, with 5 500 |
39 |
> gig drives to work with, you're right, simply making a raid1 out of |
40 |
> everything isn't possible. |
41 |
> |
42 |
> You could do a 4-drive raid10, two-way striped and two-way mirrored, for |
43 |
> a TB of storage for the media files and possibly squeeze the VMs between |
44 |
> the SSD and the raid, with the 5th half-TB as a backup, but it'd be quite |
45 |
> tight and non-optimal, plus losing the wrong two drives on the raid10 |
46 |
> would put it out of commission so you'd have only one-drive-loss- |
47 |
> tolerance there. |
48 |
> |
49 |
> You could buy a sixth half-TB and try either three-way-striping and two- |
50 |
> way mirroring for the same one-drive-loss tolerance but a good 1.5 TB (3- |
51 |
> way half-TB stripe) space, giving you plenty of space and thruput speed |
52 |
> but at the cost of only single-drive-loss-tolerance. |
53 |
> |
54 |
> You could use the same six in a raid10 with the reverse configuration, |
55 |
> two-way-stripe three-way-mirror, for better loss-of-two-tolerance but at |
56 |
> only a TB of space and have the same squeeze as the 4-way raid10 (but now |
57 |
> without the extra drive for backup), or... |
58 |
> |
59 |
> Personally, I'd probably be intensely motivated enough to try the 2-way- |
60 |
> stripe 3-way-mirror 6-drive raid10, squeezing the media space as |
61 |
> necessary to do it (maybe by using external drives for what wouldn't |
62 |
> fit), but that's still a compromise... and includes buying that sixth |
63 |
> drive. |
64 |
> |
65 |
> So the raid6 might well be the best alternative you have, given the data |
66 |
> size AND physical device size constraints. |
67 |
> |
68 |
> But some time testing the performance of different configs and |
69 |
> familiarizing yourself with the options and operation, as you've decided |
70 |
> to do now, certainly won't hurt. I DID say I wasn't real strong on the |
71 |
> chunk options, etc, myself, and you're using ext4, not the reiserfs I was |
72 |
> using, and I believe ext4 has at least some potential performance upside |
73 |
> compared to reiserfs, so it's quite possible that with some chunk/stride/ |
74 |
> etc tweaking, you can get something better, performance-wise. Tho I |
75 |
> expect raid6 will never be a speed demon, and may well never perform as |
76 |
> you had originally expected/hoped. But better than the initial results |
77 |
> should be possible, hopefully, and familiarizing yourself with things |
78 |
> while experimenting has benefits of its own, so that's an idea I can |
79 |
> agree with 100%. =:^) |
80 |
> |
81 |
> -- |
82 |
> Duncan - List replies preferred. No HTML msgs. |
83 |
> "Every nonfree program has a lord, a master -- |
84 |
> and if you use the program, he is your master." Richard Stallman |
85 |
> |
86 |
> |