Slides from May talk now available

[Lennart Sorensen]

On Tue, May 19, 2009 at 06:23:35PM -0400, S P Arif Sahari Wibowo wrote:
> Thanks Rob! Make feel more unfortunate not be able to come. :-)
>
> Few questions I have in mind; sorry if these questions already came up in 
> the talk before.
>
> For home purpose and probably small businesses, they may want to take a 
> look at other solutions which are cheaper or at least less overhead cost:
>
>  * Archival grade (so called medical grade) CD-R and DVD-R. Yes we know 
> that "100 years CD" is just myth and that many cheap CD-R & DVD-R does 
> not even last a year; but how about this archival grade stuff (and where 
> we can get it in Canada, anyway)? With good error correction mechanism 
> (any good open source solution?) can this last to - say - 10 years 
> reliably?

Package: dvdisaster
Priority: optional
Section: otherosfs
Installed-Size: 832
Maintainer: Daniel Baumann <daniel-8fiUuRrzOP0dnm+yROfE0A at public.gmane.org>
Architecture: i386
Version: 0.70.3-2
Depends: libatk1.0-0 (>= 1.12.2), libc6 (>= 2.3.6-6), libcairo2 (>= 1.2.4), libfontconfig1 (>= 2.4.0), libglib2.0-0 (>= 2.12.0), libgtk2.0-0 (>= 2.8.0), libpango1.0-0 (>= 1.14.8), libx11-6, libxcursor1 (>> 1.1.2), libxext6, libxfixes3 (>= 1:4.0.1), libxi6, libxinerama1,libxrandr2, libxrender1
Recommends: dvdisaster-doc
Filename: pool/main/d/dvdisaster/dvdisaster_0.70.3-2_i386.deb
Size: 274046
MD5sum: 3351bfe1e609942f5839e893ebaefebc
SHA1: c0b243812258375ea600b048a1a2d960d2fecf97
SHA256: 34c7604fc8a768ca94d08bb36cbdbf61c427951670b3d9dc438294b972de5b48
Description: data loss/scratch/aging protection for CD/DVD media
 dvdisaster provides a margin of safety against data loss on CD and DVD media
 caused by scratches or aging media. It creates error correction data which is
 used to recover unreadable sectors if the disc becomes damaged at a later time.
Tag: uitoolkit::gtk

You can span an archive across many discs and spread the error correction
data too, so that it can even recover from complete loss of one or more
discs in the set.

>  * Solid state drive / storage. Sure if you scale it, it will be  
> expensive. But if you don't need that much data to backup, it can be 
> cheaper than tape, since the low overhead. And its longevity seems 
> promising, considering no moving parts; certainly seems better than hard 
> disk. And when you said to take a look in 5 years, the cost most likely 
> will be down much more by then.
>
> And several caveat I think should be point out:
>
>  * With tape, except for few (or just LTO?) open standard, each  
> manufacturer have their own tape format; so there is a risk - especially 
> for long term backup - that your tapes ended up unreadable if the 
> manufacturer stop producing the drive. Might need to chose "winning" 
> brand.

I think LTO is that choice these days.  I believe sony has announced
recently that they are discontinuing development of AIT (and SAIT)
and switching to LTO as well.

LTO always promisses that drives can write the previous generation and
read the two previous generations (or more), and each generation seems
to come out every 3 years or so.  Being linear also means much less wear
on the tape than helical scan tapes.  They are also very fast and seem
very reliable.

>  * Is there any study on hard disk data longevity when it just sitting 
> not turned on for long time? If not, then it may only viable for 
> "connected" backup.

Good question.

> And about your presentation:
>
>  * Why "Genealogy", "example scheme", and "what can we recover" under the 
> subsection "Tape vs Disk"? This seems to be more appropriate under 
> "What?" or "When?".

Well both tape and disk are magnetic so they probably have similar issues
although the tape media could degrade over time perhaps, while bearings
in the disk could seize.  Who knows.  I think the real answer is that
you always keep at least 3 copies, and every 5 to 10 years you transfer
everything to new technology so that your archives are never very old.
At least with digital data you can do that, with analog that was much
harder.

-- 
Len Sorensen
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists