Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 19 Aug 2004 17:07:32 +0930
From:      Greg 'groggy' Lehey <grog@FreeBSD.org>
To:        Poul-Henning Kamp <phk@phk.freebsd.dk>
Cc:        Wilko Bulte <wb@freebie.xs4all.nl>
Subject:   Re: RAID-3?
Message-ID:  <20040819073732.GV85432@wantadilla.lemis.com>
In-Reply-To: <12437.1092899768@critter.freebsd.dk>
References:  <20040819070716.GS85432@wantadilla.lemis.com> <12437.1092899768@critter.freebsd.dk>

next in thread | previous in thread | raw e-mail | index | archive | help

--thwsKKN5whlRGe6j
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Thursday, 19 August 2004 at  9:16:08 +0200, Poul-Henning Kamp wrote:
> In message <20040819070716.GS85432@wantadilla.lemis.com>, "Greg 'groggy' Lehey"
>  writes:
>
>>> Every write takes exactly the same amount of time.
>>
>> Which, including aggregate seek time, is longer than for RAID-5,
>> because more disks are involved.
>
> RAID3 is within epsilon of the single disk because all the disks
> work in unison.

Exactly the point I'm trying to make.  For multiple transfers, all
other RAID forms are faster than a single disk.

> (Spindle-sync is a good idea btw).

It can't harm.  But how much difference does it make with modern
cached drive?

>>> There is no waiting for data to be read off of any disks.
>>
>> Sure there is.  There's always waiting for data to be read off disks.
>> That's part of the way disks are built.  You've got to seek first,
>> then you've got to get the head over the data.  That's why I said that
>> RAID-3 is only useful for sequential transfers.
>
> You're wrong.  RAID-3 is good for normal usage.  The point is that
> you don't have to do the complicated "I have disk1 in my cache but
> that is not the parity and not the one I'm writing so I need to
> read 2,3 and the parity which is 4 and then write my data to disk
> 5 and calculate and update the parity on 4" dance.

Agreed.  Easier on the programmer.  The user pays.

> RAID3 works by:
>
>     A write-request:
> 	...
>
>     A read-request:
> 	...

Yes, I know all this, once we've accepted that we are no longer
dealing with the real RAID-3, but a modified version.  It has in
common with the original RAID-3 that while we're doing this, the other
requests are waiting for us to get our act together.

> 	read parity from disk5 and check

Really check every time?   I would have thought you'd only use the
parity disk if one of the other ones had gone down.

> And that is _all_ there is to it.

"What you see is all you get".

>> Note, of course, that RAID-5 is relatively good on reading.  The big
>> disadvantage of RAID-5 is when you write.
>
> RAID3 doesn't suffer and is very predictable in either mode.

As I pointed out earlier, predictability and performance are two different
kettles of fish.

>> Of course it has.  Once you spread your data out over more than one
>> disk, you need some kind of mapping.  What we're talking about here
>> appears to be an implicit one sector stripe size, though the original
>> paper talked of a stripe size of one byte.
>
> Forget the original paper.

Why?

> Original papers are full of things people have not found out yet.

I'm not sure how to interpret this.

Greg
--
Note: I discard all HTML mail unseen.
Finger grog@FreeBSD.org for PGP public key.
See complete headers for address and phone numbers.

--thwsKKN5whlRGe6j
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.0 (FreeBSD)

iD8DBQFBJFi8IubykFB6QiMRAno0AJ9pxI/c5bluqijVcfngSrC9hcKV4QCfSZEj
78rxXP8+08bIe7r+VBX6Prc=
=GzMZ
-----END PGP SIGNATURE-----

--thwsKKN5whlRGe6j--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040819073732.GV85432>