As trolls go, he falls into the same trap as dribble though - he ends up writing a significant quantity of the posts himself... the really expert trollers only need light the blue touch paper once and stand back ;-)
g6k$o63$ snipped-for-privacy@news.albasani.net...
And take out all the tape drives in a tape based sysatem. How are you going to read the tapes then?
MBQ
It's cos we've got so much time on our hands from not having to shuffle backup tapes around.
MBQ
Yeah. That would suck. Would have to failover to the second data centre that's miles away. Of course, wouldn't need tapes, as the data would already be there (replicated, on disks).
A problem large enough to take out both sites would be enough of an issue that I suspect I'd be taken out as well so wouldn't care :)
And for a home system, not convinced tapes are realistic either.
Now, back to angle grinders and WD40. I come here for something I enjoy not for more work related stuff :)
Darren
Finally you get the point/.
The point at which you start being prissy about data recovery is when its the least of your problems.
Its ALWAYS a cost benefit equation.
Once upon a time some years back some twit in a suit was trying to sell key man insurance to my business partner, 'in case I got run over by a bus'
'How much will that cost?'
He named a staggering sum.
'The only way to make that pay is if you PUSH me under the bus immediately' I said ' the business will be dead if we pay those sorts of premiums, whether I am or not'
He wasn't QUITE as thick as Dennis, fortunately. WE took the risk didn't pay the premium and ended up with a couple of million. Fair risk reward really.
>
I never said they were in case you think I did.
Congratulations. You've just invented XML. :o)
I used to back up to DC6250, then 4mm DAT, then 8mm DAT, but as disks got bigger, this became impractical (and too expensive) so I switched to exchangeable disks about a year ago. Works a treat.
yes. That's pretty much the way its been with me too.
new setup. research backup methodologies, do sums, decide ' second disk' fulfils spec at lowers price and least manual intervention.
most computer cost seems to be in the screens, motherboard and the CASE. RAM and disk is almost trivial.
Backup devices derived from media applications tend to be poor.
- DAT was good until you needed it, particularly inter-drive variability re helical scan calibration.
- CD is good subject to manufacturer, freezer can help readibility problems.
- DVD is good subject to manufacturer, freezer can help readibility problems.
- DVD-RAM should be better because it is based on Phase Change Magneto Optical with hardware verify after write. Unfortunately it is ridiculously slow to write and from media use the discs do "get stuffed" with heavy use.
Pressed CD DVD are different to your drive; the early writers had lasers that overheated and cross-readibility could be an issue. One reason I'm not jumping at Blu-Ray backup devices re infant technology; every technology has its sweet spot where media has sufficient pricing power to (almost) guarantee some QC and devices are stable in terms of cross-readibility inter-brand and intra-brand (unlike lovable DAT, might as well been /DEV/NULL).
DLT, Digital Linear Tape, was at least designed for computer applications and the price confirmed it.
The result is people use hard drives for most backup applications.
Backup is a function of many parameters...
- Ease of regeneration - if you can't regenerate it is critical to archive, test backups, offsite
- Time for backup - dedicated backup network, not backing up that which can be regenerated
- Time for recovery - too often TB are backed up yet can not be restored in an economic time
- Need for availability - RAID is about availability, which also means spare RAID card h/w & s/w revision!
- Migration with technology - Moores law applies to cost of backup, Dilbert applies to keeping backup media past technology obsolescence (3.5" Magneto Optical, Wangtek Tape).
RAID is not a backup device; online exposes you to risks of PSU failure to bad data mirroring (a brilliantly common result of RAID).
Online backup has the benefit of easy validation, offline backup has the benefit of removing accidental deletion - of course dropping a backup hard drive highlights the single point of failure problem.
Due to size of data-sets these days vs backup devices, online backup media are favoured. Mirrored servers at geographically disperse locations over VPN tunnel or dedicated comms. HD have the benefit of "bulk, cheap, fast-online" solving the long running problem of backup recovery which is the backup device is slower than the original HD media with the world operating on "HD media time". Even so, plenty of Data Centre burnt finding data recovery D2D takes too long, hence transaction-based server systems and availability is as important as "backup" these days. That requires backing-up strategy matched to how often the data changes, you don't backup static data to media that has slow recovery time. Online backup servers with automatic failover matters.
The biggest problem about backups is still human error; not making the backup, not testing the backup, right thro to the management of companies making the media - cutting corner on substrate chemistry (early TDK India CD) or moving production to a new country and using initial production run "bombs" as production candidates when they are merely landfill. Samsung & WD improved their reputation over Seagate merely because their management were not quite as greedy as Seagate America. A good HD today may be trash tomorrow due to design fault, production fault, usage fault or human error.
I use a laptop, backup to DVD-RAM when I can remember, USB more often, but another identical laptop much more often. If this one throws a wobbly I just close the lid and open another, for me its about continuity. I buy used corporate grade and turnover recycling the capital, not too bothered about the latest whizzo or particularly graphics. Biggest problem in laptop land is as much the few SMT plants
- there are about 5 and only 2 are any good. Cutting corners reduces cycle time and gets more customers locked in against competitors, that machines fail sooner is next years problem if at all.
So mix HD brand, backup type, and treat data as "paper based in a rented crap office subject to flooding from whoever above". Just make sure you do change with technology, that does not mean early adopter - it means not archiving stuff on media where the backup drive is an antique. There is a vast amount of "yeah, still got the backup... no device to read it tho". Junk devices are just that - remember the Jumbo drives, punched paper tape had at least some efficacy albeit slow in comparison! Not much better than the audio cassette backup (one reason I do not trust media based backup technology and why much of the world doesn't either - HD backups can be online, offline, remote, rapid recovery, rapidly made).
Real problem with backups is human - they don't get made :-)
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.