From: Adam Goryachev <mailinglists@websitemanagers.com.au>
To: David Brown <david.brown@hesbynett.no>,
stan@hardwarefreak.com, James Plank <plank@cs.utk.edu>,
Ric Wheeler <rwheeler@redhat.com>
Cc: Andrea Mazzoleni <amadvance@gmail.com>,
"H. Peter Anvin" <hpa@zytor.com>,
linux-raid@vger.kernel.org, linux-btrfs@vger.kernel.org,
David Smith <creamyfish@gmail.com>
Subject: Re: Triple parity and beyond
Date: Thu, 21 Nov 2013 20:54:54 +1100 [thread overview]
Message-ID: <528DD86E.8040608@websitemanagers.com.au> (raw)
In-Reply-To: <528DCD36.1060808@hesbynett.no>
On 21/11/13 20:07, David Brown wrote:
> I can see plenty of reasons why raid15 might be a good idea, and even
> raid16 for 5 disk redundancy, compared to multi-parity sets. However,
> it costs a lot in disk space. For example, with 20 disks at 1 TB each,
> you can have:
>
> raid5 = 19TB, 1 disk redundancy
> raid6 = 18TB, 2 disk redundancy
> raid6.3 = 17TB, 3 disk redundancy
> raid6.4 = 16TB, 4 disk redundancy
> raid6.5 = 15TB, 5 disk redundancy
>
> raid10 = 10TB, 1 disk redundancy
> raid15 = 8TB, 3 disk redundancy
> raid16 = 6TB, 5 disk redundancy
>
>
> That's a very significant difference.
>
> Implementing 3+ parity does not stop people using raid15, or similar
> schemes - it just adds more choice to let people optimise according to
> their needs.
BTW, as far as strange RAID type options to try and get around problems
with failed disks, before I learned about timeout mismatches, I was
pretty worried when my 5 disk RAID5 kept falling apart and losing a
random member, then adding the failed disk back would work perfectly. To
help me feel better about this, I used 5 x 500GB drives in RAID5 and
then used the RAID5 + 1 x 2TB drive in RAID1, meaning I could afford to
lose any two disks without losing data. Of course, now I know RAID6
might have been a better choice, or even simply 2 x 2TB drives in RAID1 :)
In any case, I'm not sure I understand the concern with RAID 7.X (as it
is being called, where X > 2). Certainly you will need to make 1
computation for each stripe being written, for each value of X, so RAID
7.5 with 5 disk redundancy means 5 calculations for each stripe being
written. However, given that drives are getting bigger every year, did
we forget that we are also getting faster CPU and also more cores in a
single "CPU package"?
On a pure storage server, the CPU would normally have nothing to do,
except a little interrupt handling, it is just shuffling bytes around.
Of course, if you need RAID7.5 then you probably have a dedicated
storage server, so I don't see the problem with using the CPU to do all
the calculations.
Of course, if you are asking about carbon emissions, and cooling costs
in the data center, this could (on a global scale) have a significant
impact, so maybe it is a bad idea after all :)
Regards,
Adam
--
Adam Goryachev
Website Managers
www.websitemanagers.com.au
next prev parent reply other threads:[~2013-11-21 10:01 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-11-18 22:08 Triple parity and beyond Andrea Mazzoleni
2013-11-18 22:12 ` H. Peter Anvin
2013-11-18 22:35 ` Andrea Mazzoleni
2013-11-18 23:25 ` H. Peter Anvin
2013-11-19 10:16 ` David Brown
2013-11-19 17:36 ` Andrea Mazzoleni
2013-11-19 22:51 ` Drew
2013-11-20 0:54 ` Chris Murphy
2013-11-20 1:23 ` John Williams
2013-11-20 10:35 ` David Brown
2013-11-20 10:31 ` David Brown
2013-11-20 18:09 ` John Williams
2013-11-20 18:44 ` Andrea Mazzoleni
2013-11-21 6:15 ` Stan Hoeppner
2013-11-21 8:32 ` David Brown
2013-11-20 18:34 ` Andrea Mazzoleni
2013-11-20 18:43 ` H. Peter Anvin
2013-11-20 18:56 ` Andrea Mazzoleni
2013-11-20 18:59 ` H. Peter Anvin
2013-11-20 21:21 ` Andrea Mazzoleni
2013-11-20 19:00 ` H. Peter Anvin
2013-11-20 21:04 ` Andrea Mazzoleni
2013-11-20 21:06 ` H. Peter Anvin
2013-11-21 8:36 ` David Brown
2013-11-19 17:28 ` Andrea Mazzoleni
2013-11-19 20:29 ` Ric Wheeler
2013-11-20 16:16 ` James Plank
2013-11-20 19:05 ` Andrea Mazzoleni
2013-11-20 19:10 ` H. Peter Anvin
2013-11-20 20:30 ` James Plank
2013-11-20 21:23 ` Andrea Mazzoleni
2013-11-27 2:50 ` ronnie sahlberg
2013-11-20 21:28 ` H. Peter Anvin
2013-11-21 1:28 ` Stan Hoeppner
2013-11-21 2:46 ` John Williams
2013-11-21 6:52 ` Stan Hoeppner
2013-11-21 7:05 ` John Williams
2013-11-21 22:57 ` Stan Hoeppner
2013-11-21 23:38 ` John Williams
2013-11-22 9:35 ` Stan Hoeppner
2013-11-22 15:01 ` John Williams
2013-11-22 22:28 ` Stan Hoeppner
2013-11-22 23:07 ` NeilBrown
2013-11-23 3:46 ` Stan Hoeppner
2013-11-23 5:04 ` NeilBrown
2013-11-23 5:34 ` John Williams
2013-11-23 7:12 ` NeilBrown
2013-11-24 4:03 ` Stan Hoeppner
2013-11-24 5:14 ` John Williams
2013-11-24 21:13 ` Stan Hoeppner
2013-11-24 23:28 ` Rudy Zijlstra
[not found] ` <l6u3h9$l72$2@ger.gmane.org>
2013-11-25 2:04 ` Stan Hoeppner
2013-11-25 9:15 ` David Brown
2013-11-24 5:19 ` Russell Coker
2013-11-24 21:44 ` Stan Hoeppner
2013-11-24 22:31 ` Mark Knecht
2013-11-25 2:14 ` Russell Coker
2013-11-25 9:20 ` David Brown
2013-11-21 8:08 ` joystick
2013-11-22 0:30 ` Stan Hoeppner
2013-11-22 0:33 ` H. Peter Anvin
2013-11-22 0:45 ` David Brown
2013-11-21 9:07 ` David Brown
2013-11-21 9:54 ` Adam Goryachev [this message]
2013-11-21 10:32 ` David Brown
2013-11-22 8:12 ` Russell Coker
2013-11-25 18:23 ` Pasi Kärkkäinen
2013-11-22 8:13 ` Stan Hoeppner
2013-11-22 13:15 ` David Brown
2013-11-22 16:07 ` Stan Hoeppner
2013-11-22 22:59 ` NeilBrown
2013-11-23 17:39 ` David Brown
2013-11-22 16:50 ` Mark Knecht
2013-11-22 19:51 ` Duncan
2013-11-22 8:38 ` Stan Hoeppner
2013-11-22 13:24 ` David Brown
2013-11-28 7:16 ` Stan Hoeppner
2013-11-28 7:36 ` Russell Coker
2013-11-28 9:56 ` David Brown
2013-11-21 19:56 ` Piergiorgio Sartor
2013-11-19 18:12 ` Piergiorgio Sartor
2013-11-20 10:44 ` David Brown
2013-11-20 21:59 ` Piergiorgio Sartor
2013-11-21 10:13 ` David Brown
2013-11-21 17:37 ` Goffredo Baroncelli
2013-11-21 20:05 ` Piergiorgio Sartor
2013-11-21 20:31 ` David Brown
2013-11-21 20:52 ` Piergiorgio Sartor
2013-11-22 0:32 ` David Brown
2013-11-22 20:32 ` Piergiorgio Sartor
2013-11-26 18:10 ` joystick
2013-11-20 21:38 ` Andrea Mazzoleni
2013-11-20 22:29 ` Piergiorgio Sartor
2013-11-23 7:55 ` Andrea Mazzoleni
2013-11-23 22:10 ` Piergiorgio Sartor
2013-11-24 9:39 ` Andrea Mazzoleni
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=528DD86E.8040608@websitemanagers.com.au \
--to=mailinglists@websitemanagers.com.au \
--cc=amadvance@gmail.com \
--cc=creamyfish@gmail.com \
--cc=david.brown@hesbynett.no \
--cc=hpa@zytor.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=linux-raid@vger.kernel.org \
--cc=plank@cs.utk.edu \
--cc=rwheeler@redhat.com \
--cc=stan@hardwarefreak.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).