Skip to contentSkip to conversation

4k drives: benefits and models?

DISCUSSION Michael Monnerie 3 participants 3 messages Jul 26, 2010 1 message not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

#1 lore raw
Quoted 8 lines · On Montag, 26. Juli 2010 Louis-David Mitterrand wrote:

- what are the benefits of these new drives? Speed, temperature, reliability?

- what specific HD models are truely 4k for purchase right now? I'm on the market for a bunch of 2Tb drives, what is a good choice?

- what options do I need when creating my raid6 array and xfs filesystem to fully take advantage of 4k sectors?

ATM we are in a transition phase - there are first drives available, and it will take some time until every hardware and software piece is fully able to support 4K sectors. Probably all hard disk manufacturers will switch to 4K sectors within the next years, as the advantage is that more data can be saved onto a hard drive, and speed slightly increases (when used correctly). But that will be a long transition.

It's not necessary to buy 4k sector drives, I'd only recommend it if you are a techie and want to play around with it. Stay at normal drives, that's proven technology and works.

For a hardware RAID-6 controller, it shouldn't matter, as the stripe size is usually 16k or higher, which is 32 sectors (or 8x 4K sectors), and as a multiple of the sector size there's no speed difference to normal drives.

I use my 2 4k drives with a software RAID-0 in Linux, it's mostly to play around (sometimes I'm still a techie), and get an idea of how good the technology can be used already.

I have no idea about Windows support for 4k sector drives.

Signature · 13 lines
mit freundlichen Grüssen,
Michael Monnerie, Ing. BSc

it-management Internet Services
http://proteger.at [gesprochen: Prot-e-schee]
Tel: 0660 / 415 65 31

****** Aktuelles Radiointerview! ******
http://www.it-podcast.at/aktuelle-sendung.html

// Wir haben im Moment zwei Häuser zu verkaufen:
// http://zmi.at/langegg/
// http://zmi.at/haus2009/

_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs

#2 lore raw
In reply to a message not in the archive
Quoted 3 lines

- what are the benefits of these new drives? Speed, temperature, reliability?

It is purely higher density. A longer physical sector means less percentage of the disk is "wasted" on metadata. This benefits the manufacturer mostly, so they can claim more capacity with less cost (but more credibly than the far more slimy trick of LCD manufacturers who are going to ever more extreme landscape ratios to claim a longer diagonal with less pixels).

Quoted 1 line

From the user/filesystem point iof view it is not a positive

thing, but since the 4BSD FFS introduced 4KiB blocks (a very bad idea) and intel and others followed throught with alrge 4KiB pages that has been largely irrelevant.

Also it seems that few people are aware that other device types have large sectors, as CDs have 2KiB sectors, DVDs have 32KiB, BDRs have 64KiB sectors (and both simulate 2KiB sectors), and SSDs can have 256KiB sectors (or rather "erase blocks", I think that they can read in smaller units than that).

Quoted 3 lines

- what specific HD models are truely 4k for purchase right now? I'm on the market for a bunch of 2Tb drives, what is a good choice?

These are two completely unrelated questions, and the second is sort of irrelevant because there are so few manufacturers and thus models that one has to buy all of them. Except of course for those who fill their arrays with disks from the same brand, model and even shipping carton, as they know better.

As to the first, there have been a few discussions in the past on vsarious Linux mailing lists etc, but the list of models changes as manufacturers change firmware and "truely" is thus a somewhat elastic concept dependent on revision not just model.

Quoted 2 lines

- what options do I need when creating my raid6 array and xfs filesystem to fully take advantage of 4k sectors?

Some recent posts to this mailing list give those. I would add also large inodes (2Kib), to alleviate the big downside of 4KiB sectors for small files.

_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs

#3 lore raw

Le Mon, 26 Jul 2010 10:35:39 +0100 [email protected] (Peter Grandi) écrivait:

Quoted 7 lines

It is purely higher density. A longer physical sector means less percentage of the disk is "wasted" on metadata. This benefits the manufacturer mostly, so they can claim more capacity with less cost (but more credibly than the far more slimy trick of LCD manufacturers who are going to ever more extreme landscape ratios to claim a longer diagonal with less pixels).

It isn't actually that simple. By keeping the same metadata ratio, the largest correctable error goes up from 10 bytes to 80 bytes. That means taht the new drives should be much less prone to uncorrectable ECC errors, which are the real culprit with the previous generation of big drives (Seagate barracuda 1, 1.5 and 2TB are particularly terrible IMO).

Quoted 4 lines

From the user/filesystem point iof view it is not a positive thing, but since the 4BSD FFS introduced 4KiB blocks (a very bad idea) and intel and others followed throught with alrge 4KiB pages that has been largely irrelevant.

This was eons ago, I'm pretty sure we managed to live with it :) What was problematic back when disks drives were in the hundreds of megabytes isn't so much a problem when the smallest drives available (146 GB) are 1000 times bigger. Many filesystems use much bigger blocks on bigger filesystems, too (NTFS anyone? WS2003 can use 16, 32 or even 64 KB).

Signature · 11 lines
------------------------------------------------------------------------
Emmanuel Florac     |   Direction technique
                    |   Intellique
                    |	<[email protected]>
                    |   +33 1 78 94 84 02
------------------------------------------------------------------------

_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs