From: Bill Davidsen <davidsen@tmr.com>
To: linux-raid@vger.kernel.org
Subject: Re: Sleepy drives and MD RAID 6
Date: Thu, 14 Aug 2014 14:10:15 -0400 [thread overview]
Message-ID: <53ECFB87.3010700@tmr.com> (raw)
In-Reply-To: <20140812092901.60f007e6@natsu>
Roman Mamedov wrote:
> On Mon, 11 Aug 2014 18:03:17 -0700
> Adam Talbot <ajtalbot1@gmail.com> wrote:
>
>> I need help from the Linux RAID pros.
>>
>> To make a very long story short; I have a 7 disk in a RAID 6 array. I
>> put the drives to sleep after 7 minutes of inactivity.
>
> It is well known that repeatedly spinning a drive down/up is absolutely the
> worst possible thing you can do to it, from a long term reliability standpoint.
> So my personal suggestion would be to reconsider if you really want this. The
> power consumption from 7 spinning drives with no access should be no higher
> than 60-70 watt; IMHO saving that amount, is not something that's worth risking
> your disks and data for.
>
Unless you live in someplace really cold, there's the cost of pumping that heat
out of the room. Running A/C can almost double your power cost, and running the
room hot shortens your component life. In other words there's more cost than the
power for most people. Even if my hardware can run at 90F, I can't.
There appears to be a partial solution, get a small SDD (<$100) and put the
journal on it. Get two, run RAID1 if you must. Then configure the system to
write the journal with all data (data=journal), and the write will be really
fast, even if they don't fully complete for a minute or so. Doesn't help reads,
of course. Turns the journal into cache, sort of.
I have the feeling that sequential spin is an option in a driver, but I can't
remember or quickly find where. I say this because I had to set it on one
machine I had, spinning up the whole array at one caused the power supply to
overload, and until I could get a bigger one which would fit I set an option.
That was long enough that I can't remember where I found that. Turn it off and
all seven drives will ask for power at once, which probably isn't a great thing,
but not my system.
--
Bill Davidsen <davidsen@tmr.com>
"We have more to fear from the bungling of the incompetent than from
the machinations of the wicked." - from Slashdot
next prev parent reply other threads:[~2014-08-14 18:10 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-08-12 1:03 Sleepy drives and MD RAID 6 Adam Talbot
2014-08-12 1:21 ` Larkin Lowrey
2014-08-12 1:31 ` Adam Talbot
2014-08-12 5:55 ` Can Jeuleers
2014-08-12 9:46 ` Wilson, Jonathan
2014-08-12 15:26 ` Adam Talbot
[not found] ` <CAH_2GhfBKiJAT=+zCFH0_xkmyTfWFi=c-F+z8rwL1x++UwU+KA@mail.gmail.com>
2014-08-13 16:07 ` Adam Talbot
2014-08-14 16:50 ` Adam Talbot
2014-08-14 17:00 ` Larkin Lowrey
2014-08-14 17:37 ` Adam Talbot
2014-08-14 18:05 ` Larkin Lowrey
2014-08-18 15:41 ` Adam Talbot
2014-08-18 16:16 ` Larkin Lowrey
[not found] ` <CAH_2GhcwHszr75bbDq2aWRnnF8-ttpGERKE=CKt=Nri3ue9row@mail.gmail.com>
2014-08-28 21:04 ` Adam Talbot
2014-08-12 3:29 ` Roman Mamedov
2014-08-14 18:10 ` Bill Davidsen [this message]
2014-08-14 19:33 ` Adam Talbot
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=53ECFB87.3010700@tmr.com \
--to=davidsen@tmr.com \
--cc=linux-raid@vger.kernel.org \
/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