Selling ZFS

Quicksearch

Disclaimer

The individual owning this blog works for Oracle in Germany. The opinions expressed here are his own, are not necessarily reviewed in advance by anyone but the individual author, and neither Oracle nor any other party necessarily agrees with them.

Navigation

< Eigener Planet? | 20 Jahre >

Wednesday, November 1. 2006

I was at a customer site today to talk about storage and NAS. And again the usual problem. There is a catch in selling ZFS to customers. When you tell them of the RAID-5 write hole or silent data corruption they ask you: "Do you tell me that my data isn´t secure on your products without ZFS?". When you say "yes" you discredited you current product line. When you say "no", they don´t see the need for thinking about such topics like verifiable data integrity and mirror self healing.

The problem is even worse than you think at the first time: We are the only company talking about a complete verifiable file system and the problems inherent in conventional filesystems/volumes. To many customers see this as a sun-only problem but it´s an industry-wide problem of file storage on rotating rust. It´s difficult to deeducate you customers from the belief that data on disk is immutable unless you willingly change it without set him in panic mode.

Posted by Joerg Moellenkamp in English, Solaris at 19:58 | Comments (4) | Trackbacks (0)

Defined tags for this entry: solaris

, storage

, sun

, zfs

Related entries by tags:

View as PDF: This entry | This month | Full blog

Trackbacks

Trackback specific URI for this entry

No Trackbacks

Comments

Display comments as (Linear | Threaded)

Maybe to calm the customer down a little bit, is to tell him, that statistically this was until now, not a huge problem. But with ever growing disks, the chance for silent corruption will grow. (Of course the problem has existed always...)

#1 Mika on 2006-11-02 16:39 (Reply)

But it´s like the czar who kills the herald of a bad message. I use the same argument: Hard disks get so huge, that you you have a bit error when you you´ve read the disk 10 or 20 times ...

#1.1 Joerg (Homepage) on 2006-11-02 18:22 (Reply)

I really hate RAID-5 and tell that everybody, nobody wants to listen.

Some know about the weakness of RAID-5 and tell you there are some solutions. I yet havent seen any controllers who do good scrubbing without software support, and I am quite sure that I/O errors and block errors, especially multi disk errors are an unsolved problem.

Bernd

#2 Bernd Eckenfels (Homepage) on 2006-11-04 06:58 (Reply)

I don´t think that scrubbing should be part of the controller. You need to much intimate knowledge of the filesystem to really scrub a filesystem. Or you end with an controller that has knowledge of the logical structure of the filesystem.

But i think there is an solution for the problems you´ve described, and the solution is called ZFS.

#2.1 Joerg (Homepage) on 2006-11-04 07:31 (Reply)

Add Comment

Name
Email
Homepage
In reply to
Comment	E-Mail addresses will not be displayed and will only be used for E-Mail notifications. To prevent automated Bots from commentspamming, please enter the string you see in the image below in the appropriate input box. Your comment will only be submitted if the strings match. Please ensure that your browser supports and accepts cookies, or your comment cannot be verified correctly. Enter the string from the spam-prevention image above: Standard emoticons like :-) and ;-) are converted to images. Enclosing asterisks marks text as bold (word), underscore are made via _word_.
	Remember Information? Subscribe to this entry