
Hard Drives Instead of Tapes? 484
An anonymous reader writes "Tom's Hardware News weekly news letter has a very interesting article about Dr. Koch of Computertechnik AG who won the contract to build a RAID backup system for the University of Tübingen. Dr. Koch took several standard entry-level servers, such as the dual-Athlon MP, and add modern components and three large-caliber IDE-RAID controllers per computer, and a total of 576 x 160GB Drives."
Far more practical (Score:5, Interesting)
This is a much better solution than tape, really. It's predictable that the industry will probably move in this direction, now that the hardware is cheap enough and of high enough capacity to serve this function.
Imagine: instant recovery. Your backup could be a usable image of your live server.
Re:Far more practical (Score:5, Informative)
Yeah, the full usable image would be nice, but would probably require a shutdown for data consistency. The backup strategy would likely be similar to that of an Oracle system cold backup.
-Alex
Re:Far more practical (Score:5, Insightful)
Re:Far more practical (Score:4, Informative)
My understanding was that for tape it is only 5-10 years, but that could very well be out-dated. What is the current shelf life for magnetic tape?
Re:Far more practical (Score:3, Insightful)
Re:Far more practical (Score:4, Funny)
"What's that clicking sound?"
Re:Far more practical (Score:3, Insightful)
OTOH, my father has 40-year-old punch cards that read just fine. (Course, that doesn't scale to terabytes.)
The upshot is that for long-term (>10 years) backup, have a refresh plan in place, where the data is periodically verified, and if necessary, extracted and copied to fresh media. (I have some
Re:Far more practical (Score:3, Interesting)
And if you're really paranoid, each one has it's own power source, etc, etc, etc.
Re:Far more practical (Score:5, Interesting)
Granted, most current tape systems support more storage than current optical systems. But optical backup systems are gaining popularity. I expect that as they become more popular, they will become cheaper, more reliable, faster, easier, and more spacious... just like every other form of technology that gained popular view.
Tapes have been trust-worthy for some time now, but they are slow and bulky. As the amount of data to backup continues to increase, tape won't be able to keep up. Optical media has progressed in speed considerably over the last few years. I know that some media (CR-R's for example) have hit media thresholds... but they're still MUCH faster than tape at current speeds.
I expect that optical media will continue to improve. In my original post I did state "or something similar", to allow for progress over current optical limitations. My main point really being that tape is dying out. An optical media is becoming a focus for long term storage media, and investment will go into improving upon it.
-Alex
Re:Far more practical (Score:5, Interesting)
The benefits of having backup to disk is of course speed. But what happens when you have a disaster? Your SOL, because your backup-on-disk system just got toasted too.
The benefits of having backups on tape is that you can send those tapes anywhere. It might not be as quick as sending a file electronically, but when you are talking hundreds of gigabytes of data, it just isn't economical to do anything but tape.
Tape will never die. Hardware may be cheap and high capacity, but transmission costs keep it from being feasible.
You also need to take a look at space utilization. You can put a tape silo into a footprint that gives you much much more capacity per square foot than disk.
Re:Far more practical (Score:3, Interesting)
Re:Far more practical (Score:5, Interesting)
Re:Far more practical (Score:3, Interesting)
The perfect solution to backups would be notebook SATA disks, which should hopefully appear soon. Hotpluggable, no bulkier than an LTO or DLT tape, screaming fast compared to LTO and DLT, and very hard to dam
Compliance (Score:2, Interesting)
Re:Compliance (Score:5, Insightful)
Oh PLEASE! I worked for, what was at the time, the 17th largest CC processor in the nation. Not so big, but lots of merchants. They bought a front-end (where your credit card terminals dial into), and built a backend settlement (so they didn't need FDR - who recently ROYALLY hosed everyone with a software update, including CHASE themselves. No, this software update was completely seperate from the SQL Slammer worm that took them down when it appeared.).
Complaince, usually done by the OTS (Office of Thrift Supervision), is NOT ISO 9000 type stuff. Financial companies are CHEAP. Never forget that. Whatever is the cheapest solution, is the one that is used.
As for tape backups - as an example: It took quite a bit of convincing to upgrade from the 4 drives that took two days to backup the whole network to a single Sont DLT drive. (Because $70/tape is a LOT of money)
There were no 'compliance' worries at all.
Re:Compliance (Score:3)
That's not the financial services industry in any meaningful sense of the word, just a teeny tiny corner of it. If you're trading real instruments - think Wall Street, or the Square Mile - you need to keep everything around for 7+ years, and if anyone you've traded with in that time gets audited, you might be asked for your counterparty records. Not to mention the fact that you can trade instruments with a maturity
Re:Compliance (Score:2, Insightful)
They of course are also important for business continuity, as Sept. 11, 2001 showed us when several large finacial firms had their data centers destroyed.
Sound fine, but... (Score:2, Insightful)
I mean, sure tape isn't great, but it's a lot more transportable than harddrives.
Re:Sound fine, but... (Score:5, Informative)
Re:Sound fine, but... (Score:3, Interesting)
Re:Sound fine, but... (Score:5, Informative)
Re:Sound fine, but... (Score:5, Informative)
All systems have live fail-overs. When not required by law, and they frequently are, such systems are required by the demands of profit. If financial transactions falter for a *second*, it means money lost.
Back-up media is triple redundant and incremental over 5 days. Backup irregularities of any kind are logged, investigated, and acted upon by at least 3 individuals.
Copies of backups are stored both on site and off-site in a secure location provided by our insurance provider. We make frequent trips to this secure location daily in order to deposit backups. These procedures are audited and reviewed on a regular basis by both internal auditors and regulatory board auditors.
Tape is just a little more reliable than IDE in this kind of situation. Tape is going to be more recoverable, even in case of a long drop or serious auto accident between point A and point B. If necessary, teap will also survive shipping better.
Sorry, guys. As reliable as IDE drives have become, they're just not as durable as a tape cartridge. With the sheer amount of backup we keep, it's also significantly cheaper.
Re:Sound fine, but... (Score:4, Insightful)
Not all companies need five 9's. Not all companies lose much money if data or systems are not available for a short time. In fact, I'd say it's the majority of companies that fall into that category.
Extreme reliability and availability are extremely expensive. For most companies, it's not worth it.
I agree with you, Large ATA RAID probably isn't for your industry, it's not right for everyone. It does work fine for lots of people though. I expect to see it cover much of the 5TB range of near-line backups in the next few years.
Re:Sound fine, but... (Score:3, Interesting)
Re:Sound fine, but... (Score:2)
I could envision a super-fat pipe being used to mirror a facility to a neighboring (or even geographically-distant) facility along with a system like this.
rdiff-backup saved me when a power supply blew out on a server. Within an hour of the failure, I was back up and running on the backup server. It could have been much faster had I automated the failover...
Re:Sound fine, but... (Score:2, Interesting)
But wait there's more. Those mag tape cartriges have a transfer rate of about 10 MB/sec. With hard drives, your backups will take a fraction of the time they took under the old system. That leaves plenty of extra time to pack the drives up extra securely. You may even be
Re:Sound fine, but... (Score:5, Informative)
Uh, I think you better look at tapes again. AIT-3 is 100GB uncompressed. Super-AIT is 500GB uncompressed. Transfer rates for Super-AIT are in the 30 GB/s range uncompressed. All of these numbers go up with compression, which is built into the tape drive hardware -- assuming you're storing compressable data.
All in all, they're likely to have a higher sustained transfer rate than IDE drives, and are going to be more reliable, less costly in bulk, and easier to handle.
Of course they're silly for small systems... but that's not what we're talking about at all.
Re:Sound fine, but... (Score:2)
Re:Sound fine, but... (Score:3, Insightful)
Re:What about reading the article? (Score:2)
Last time I looked, a 40/80/160 gig hd wasn't all that big. I do my backup onto one, then bring it home and back THAT up to my home machine, so I end up w 4 copies of any file (my workstation, the server, the backup hd, and my home machine). Time to restore is pretty much zero (just plug the hd in, set the new pa
Re:What about reading the article? (Score:2)
I don't know if that is a good idea (Score:4, Insightful)
Re:I don't know if that is a good idea (Score:3, Funny)
Re:I don't know if that is a good idea (Score:4, Insightful)
The best backup solution would be a bunker with hard drives, backed up via fiber in real time.
Re:I don't know if that is a good idea (Score:2)
Re:I don't know if that is a good idea (Score:2)
Yeah, true - that's what I meant[g]
Optical isn't necessarily immune to EMP (Score:3, Interesting)
Re:Optical isn't necessarily immune to EMP (Score:5, Funny)
In all likelihood the entire human race has also ceased to be a concern.
Re:I don't know if that is a good idea (Score:3, Funny)
We could punch into paper tape.
Re:I don't know if that is a good idea (Score:5, Funny)
Earthenware tablets, made of clay fired at low temperatures (1816F/991C), seem to do nearly as well, while stoneware tablets, made of clay fired at high temperature (2345F/1284C), last about the same as actual stone. Ceramics have relatively high resistance to moisture and thermal variation. Depending on the clay composition and the application of glazes, there is variable resistance to acid. Ceramics do not handle physical shock particularly well.
Glass can last thousands of years, but is vulnerable to shattering or acid.
None of these, however, are earthquake-resistant. Outside of the immediate blast radius, they're good against nukes.
Etching into stainless steel is good, although in the event of a nuclear attack, this would be succeptible to melting (or self-destruction due to induced current) within a certain area. It handles thermal and moisture extremes pretty well, but doesn't handle acids well.
Stamping into gold foil is expensive, but quite durable. It's immune to some of the chemical risks posed by steel, but is more likely to be stolen. It's also not as hard, thus leading to risk of data corruption or loss via impact.
Parchment, preserved lamb or sheep skin, can last a very long time (on the order of 2,000 years) in the right conditions. It does well with exposure to electromagnetic radiation, but deals badly with moisture or excessive dryness, and is highly vulnerable to acid.
Delay-line broadcast (reflecting your data with a laser off of a distant object, and rebroadcasting ad infinitum) is fairly reliable until occlusion of the data path occurs, or the transceiver is smashed, unplugged, EMPed.
Yeah, data preservation is hard in the long haul.
Re:I don't know if that is a good idea (Score:5, Funny)
This reminds me of a formal budget proposal submitted by my predecessor many years ago (I run the IT dept. at a small college). He gave a very detailed cost breakdown of several means of replacing our then-current backup and recovery method for our file server's RAID array (we were very small way back then). He had costs for hardware, time, and manpower for just about every option available at the time.
His last option, put forth just as seriously and fully as the rest, included the cost of having a team of monks write out the data by hand onto reams of paper, bit-by-bit. Then for recovery, the monks would re-enter the data back into the computer, bit-by-bit. On the pro side he argued that monks work cheap and are very dedicated to what they do. But the con was the time involved for this method was somewhat prohibitive.
Belloc
This would work for limited installations (Score:3, Insightful)
Just remember, if you can build something like this for backup, you can also build something like this for regular storage... and then what will you do if you need to back it up? Especially if you need to have a 6 month rotating backup...
I'm afraid it will be back to tape then...
Re:This would work for limited installations (Score:2)
Re:This would work for limited installations (Score:2)
With a sufficiently powerful laser, you could encode the data and beam it off to some distant object. When the reflected beam finally reaches earth, there is your backed up data, ready for retrieval. You could pick a number of objects at varying distances to allow for longer backups of data, with reflection times of hours/days/weeks/years... Granted there would be some celestial issues that could corrupt your data, but if you picked a few redundant yet diverse backup site
Just build more (Score:2)
Data probably is corrupted much more frequently by mistakes and systems problems, and with the sort of live redundancy favored by DR architectures the b
Surprised it didn't happen sooner (Score:5, Informative)
Re:Surprised it didn't happen sooner (Score:2)
With tape, if the reader fails, you will probably still have your data, and you can just find another reader.
I would also like to point out that HD manufactures our lowering there warrenty period. IT is only a matter of time before some cost cutting measure makes them use lower quality equipment and parts.
Re:Surprised it didn't happen sooner (Score:2)
The staging solution has been available (fully automated) in Unixland for at least ten years.
Disk backups are fine for on-site backups but you still need off-site copies, which is done either by tape or over-the-wire synchronizing. The latter gets real pricey real quick.
Offsite? (Score:5, Insightful)
Maybe you could do it with a big pipe between your backup location and your servers. But I bet that would cost a bundle in bandwidth.
Also did anyone notice that typo on UPS (maybe they were on drugs USP [usp.org])! It took me a good minute to catch it.
Re:Offsite? (Score:3, Insightful)
I've been using HDDs for backup for awhile now. Tapes were just way too much hassle, too expensive, and too fragile for my daily backups. I don't have protection against fire, but the whole setup can backup 650GB (usable) of data, survive disk failure, and cost me $1500, and I built this a year and a half ago with 80GB drives. My nightly backups are fully aut
Re:Offsite? (Score:3, Interesting)
Re:Offsite? (Score:5, Interesting)
There's one aspect in which Dr. Koch's backup system can't keep up with tape solutions: storing the backup medium in another location after the backup has been completed.
The article didn't address what to do in this case. Instead, they continued:
As long as this isn't necessary, Dr. Koch's backup system offers some rather unique advantages.
Given that it's hardware-focussed, maybe one can understand this omission, but here in the real world it's still important. So, yes, what does one do if one does need offsite storage? Realistically, I think your suggestion of a big pipe is about the only way. It's hardly feasible to hotswap loads of drives for your offsite storage every morning. (Yes, I know they're using IDE, but think Promise controllers.)
The question then becomes a comparison of the cost of providing for offsite storage in this manner versus the saved cost of replacing your tape library with associated robots, etc.
However, the article also discusses (very briefly) associated costs for specialized backup administrators, delays inherent in recovering from tape backups, etc., so they're not totally unaware of the real-world issue. I suspect they may have chosen to ignore this specific issue because (i) it wasn't an issue in this case study, and (ii) examining it would've been a touch difficult.
Re:Offsite? (Score:2)
Help me understand... (Score:5, Insightful)
sPh
Re:Help me understand... (Score:2, Redundant)
torados are quite rare in Tübingen
however, a backup system like this protects you from (accidental) deletion of files and hardware failures. thats enough for many people..
Re:Help me understand... (Score:2)
Given that I live in Tornado Alley, I may be a bit sensitive on that topic!
However, three years ago I was visiting a number of my (then) employer's sites worldwide and there was a disturbing tendency for there to be a severe thunderstorm with tornado while I at a site - some in places where tornados are typically seen only once every 100 years. So don't be too sure!
But floods, large fires, or some sort of large-scale natural disaster can happen anywhere. Floodi
Re:Help me understand... (Score:2)
however, a backup system like this protects you from (accidental) deletion of files and hardware failures. thats enough for many people..
Well then the spec'd system is vastly overdesigned for THAT. There are far easier ways to solve both.
It is possible to do off-site storage with drive arrays, but you have to design it into the system. As it is, this system solves
Off Site Backup +/- (Score:4, Insightful)
Offsite backups, whether tape or disk, present some pros and cons.
Pro: offsite is safer from local disaster effects.
Con: data restoration takes longer from further away.
Pro: high bandwidth connection makes moving data quick enough.
Con: high bandwidth connections are expensive
Con: high bandwidth connections are susceptible to disaster induced interruption
Overall, though, I like the random access provided by disk drives over linear searches of tapes. In case the network connection is broken to the backup site, you can easily load a couple of terabytes on cheap IDE drives into the back of your station wagon and bring them to any site you like and the effective BW will still be pretty darn good.
If you drive your station wagon across the continental U.S loaded with 3 TB of IDE drives in 3 days then you will be running faster than T1.
safer away from local disaster access time is high when locals need restoration big net pipe to far away but disaster that kills the network pipe ? maybe hard drives can be couriered back.Re:Help me understand... (Score:3, Insightful)
Then you're probably out of business anyway, so what does it matter at that point?
Re:Help me understand... (Score:3, Interesting)
I'll reply to this message, but cover several similar points.
First, I should note that I do consider sanity checks and cost/benefit analysis when making backup/recovery plans. So I agree with many of these comments. BUT...
(1) Disasters happen more often than people expect. And they can happen to you, not just the other guy. Wildpackets almost went out of business [eweek.com] as a result of underestimating that.
(2) Being out of b
Nice (Score:2)
So how do you do this in this scenario? I hope they have fantastic fire extinguishers.
Re:Nice (Score:2)
And don't trifle me with complaints about bandwidth. Use rsync. If you're really generating so much new data every day that rsync is infeasible, then maybe you have to look at a different solution. But then tape probably isn't your answer, either.
uh huh (Score:2, Interesting)
So, I come in, hit the systems with a hammer, and you're done. A virus wipes you out. A malicous hacker, a stupid user, kernel oops that fucks up the filesystem, something. No tapes to go back to. Not to mention that old data that isn't being used can be archived off to tape, which may end up costing less in the long run that more and more hard drives - hard drives use power, remember?
I never understand how you can trust you data to ONLY be on media that is tied to the mechanism. If the mechanism
HDD Tapestream (Score:2, Funny)
Tapes are expensive, HDDs are cheap in comparison
My C64 uses tapes, I don't even see the competition between these two.
Re:HDD Tapestream (Score:2)
Carefull, your ignorance is showing.
Tape technology not keeping pace... (Score:4, Informative)
But, we now have $100 tapes that hold as much data as a $100 hard drive.
We switched over to hard drives for our backups at our (modest) server facility. Late last year we spent $2000 on a system with 600GB of RAID-5 protected storage. That holds current and historic backups, for around 6 months with our current load. We then weekly dump the current data-set off to a removable 120GB hard drive, which we take off-site.
Tapes are SO dead...
It works great.
Sean
Re:Tape technology not keeping pace... (Score:5, Insightful)
To backup a storage pool with under a couple of TB of storage, tape is indeed stupid. If what you need is truly massive amounts of storage that does not need to be accessed instantaneously, tape cannot be beat.
Re:Tape technology not keeping pace... (Score:2)
Re:Tape technology not keeping pace... (Score:2)
And what happens when your raid card gets flakey and begins to write garbage data? what if that happen on day 6 of your 7 day back up cycle?
Some larg instiution need backups in the terra byte range, and they need some data back-ups physically seperate from other data.
And they want to automated. and off site every day. Hard drive solution will not work.
Re:Tape technology not keeping pace... (Score:3, Interesting)
Escelade 4 port IDE RAID card: around 200 bucks
200 Gig Drives X 5: say 249 each, = 1245
Total Cost: 1945
And thats 1000 GB of un-raided space, so will end up being more than 600 GB raid 5.
Re:Tape technology not keeping pace... (Score:3, Informative)
I get the feeling that you understand how RAID-5 works, but your statement is misleading.
With RAID-5 it stores parity data across the array for each piece of information stored. So to store data that would fill N drives, you need N+1 drives for the array (1 drive extra for the parity info).
Adding drives won't protect your data any more (although hot standbys are nice to have). RAID-5 fails if I lose two drives at
Tape will be with us for a while yet... (Score:5, Interesting)
Yes, disk is good if you need instant access to your backup, and for small installations of under a couple of TB, using disk backups make sense, but for larger data pools, tape is far more economical.
Also, as mentioned in the article, disk is terrible if you need off-site backups. In addition, a tape library consumes far less power, takes up less space, and produces less heat than a drive array of the same capacity.
Basically, the death of tape has been predicted for years, but it hasn't happened yet.
Re:Tape will be with us for a while yet... (Score:4, Insightful)
Re:Tape will be with us for a while yet... (Score:3, Insightful)
On the down side, you need to keep spinning a disk in a RAID environment to make sure the data is still good. Drives with one-year warranties aren't designed to sit on a shelf for 5 years and be powered back on. When drives fail, the RAID takes over and rebuilds
ack! (Score:5, Insightful)
I ended up going on ebay and getting a StorageTek 9714 "Media Library" with 2 DLT 4000 drives in it. It takes a maximum of 2A of power.. (I've measured it much lower then that when the tape drives arn't in use..) This sucker will store up to 2.4 TB ( 1.2 TB uncompressed) in the 60 available tape slots..
The electricity saves more then makes up for the cost of the tapes.. (Also I expect the tapes to last approx 5-10 years.. I wouldn't expect that with the hard drives.)
--Mark
Re:ack! (Score:2)
Re:ack! (Score:2)
The failures generally happened to multiple drives within hours of each other, rendering the raid 5 invalid and corrupting tons of data.. (which of course required the restore off a more perminant media... DLT tape.)
I'm not saying they could have gotten a backup system for the same amount of money using convential DLT tapes and media libraries.. however, my experienc
Why not Quantum DX-30 (Score:5, Informative)
Re:Why not Quantum DX-30 (Score:3, Insightful)
I wouldn't want to support it... (Score:2, Insightful)
Re:I wouldn't want to support it... (Score:3, Informative)
Re:I wouldn't want to support it... (Score:3, Informative)
It's a worthwhile idea (Score:2)
My preference would be two sets of 4x160 in a RAID 5, using two Adaptec 2400 ATA RAID cards. That'd give me a formatted capacity of 2x 409 gigs. I'd want two of those systems available so I could have two fulls and two sets of incrementals on hand at any one time.
The only stumbling blocks I've found are: finding a 2 or 3U box that will accept two of the 24
They may not need offsite storage (Score:2)
Bad idea. (Score:4, Funny)
Re:Bad idea. (Score:2, Funny)
Re:Bad idea. (Score:2)
Re:Bad idea. (Score:3, Funny)
Put a large fireplace in your card storage room. That should dry up liquids before they can cause any significant harm, right?
Re:Bad idea. (Score:3, Funny)
Re:Bad idea. (Score:2)
Only if the IT admins get to see them applied to student bodies. BTW, what do University of Tübingen chicks look like anyway?
Three Words: (Score:3, Informative)
It's being done all over. Some people are using Network Appliances, some people are using Linux machines. Even Legato, a major player in the backup market supports backup 'staging' to spinning disk to decrease backup windows.
Tape vs disk backups (Score:2, Insightful)
I've been saved several times by being able to retrieve files 5 years old that the lawyers wanted. Because I had multiple layers of backup, even though I deleted the files from my system 3 years ago, I was able to retrieve the files from tape. They were worth far more to the lawyers than the cost of the whole backup system and tapes could ever have cost.
The typical disk backup setup does not support such archiving in depth. And
I didn't see someone say it yet, so (Score:2)
tape's still bigger (Score:3, Funny)
Why do this? (Score:2)
My company (Score:2, Funny)
What we do is plug a digital camera into the server, and copy everything to its flash media card inside. When we go on vacation we just take the backup "off-site" to the Bahamas.
And in the event of failure we also have a 256MB backup of the first bit of stuff on the hard drive, and a picture of the server room so we know what to order after it melts in a fire.
Let's see... (Score:3, Insightful)
Assume 5 years MTBF.
That end up being 100 Hard drive failures per year, about $10,000/yr, not counting labor.
Or 2 per week. ($200/wk), if efficient to replace then add another $100/wk for ordering, shipping, storage, replacement and disposal.
That's assuming good cooling and low usage (equivilant to an intermittant home user - which is what I expect a good backup system to get used to)
So, ignoring the cost of the initial investment, they'll be paying up to $15,000 per year to maintain this backup solution.
This is more expensive than many traditional backup methods, such as tape.
However there were a few 'gimmes'. Firstly, the array only has to last 5 years. Secondly they are using 5400rpm hard drives - much cooler. Thirdly, these hard drives have a 3 year warranty, which is better than most places will give you now.
So it's likely that the maintenance cost, in this case, is going to be low compared to the initial investment.
The real problem, then, is the tendancy to keep an old system long past its prime and original intent. Someone in the future will say, "Instead of junking the system and upgrading to new technology, let's just throw larger hard drives in there each time one fails and up the capacity. Eventually it will cost $10k or more per year, and they won't know it.
-Adam
Optical tape? (Score:3, Informative)
Still I think that this misbalance between tape prices and HDD prices cannot last.
Re:Moving data offsite (Score:2)
Re:Finally!!! (Score:2)