prostoalex writes "The Guardian takes a look at the current developments in the world of holographic storage. Despite being available in research for over 40 years, the technology is getting commercialized only now, with InPhase Technologies launching its 600 GB write-once disk and a drive this fall. What avout the price? "The first holographic products are certainly not mass-market — a 600GB disc will cost around $180 (£90), and the drive costs about $18,000. Potential users include banks, libraries, government agencies and corporations.""
With 1TB hard drives hitting the market, is it really worth spending $180 for 1 (!) optical disk and a $18k for the drive? For that money one can buy a lot of 1TB hard drives and build a RAID 0/1/5... array and have more capacity and reliability. Besides, I don't see museums or even companies running to get that drive, because if the standard goes the way of the Laserdisc [wikipedia.org] then they are stuck with some exotic technology experiment and when their drive breaks there they will not be able to easily get thei
Bleeding edge is always a ridiculous expense. The people who are willing to be there already know who they are. That you even raise this question means that you are not.
OTOH, neither am I, but that's not the point. The point is, this is the first commercial volley of a new technology, which means that a few years hence it will be cheaper with even higher data densities.
Meaning, potentially, something like the entire run of every season of every Star Trek series ever... on one disc.
The tape/HD size ratio is getting so ridiculous that at work we're seriously considering using Hard Drives as removable tape-like media for backup. Any other solution for backing up terabytes of data is too expensive.
We've been doing this for a few years actually, as a "roll your own" solution. We currently use removable drive carriers from DataStor, and 500 GB Seagate disks (first ATA, now SATA). We also use foam-padded locking carriers that are take off-site every day. We do ~1.2 TB of backups every nigh
Toss an autoloading platter jukebox onto the front of this and you can start long term archival recording of business documents (mortgage documents > 30 years).
You obviously responded without reading my whole comment (tsk...tsk!).
Alright, so say you buy the $18,000 drive and 200 discs as $180 a piece. You spend a couple of months saving all your highly valuable data, put it in a vault and wait a 30 years. BUT, the next year, the company that created the $18,000 and their proprietary storage goes bell
Yeah and at $18,000 for a drive these big companies don't have to worry about people losing discs anymore because they'll be the only ones who can afford the drives to read them!
I remember the very first DVD writer which was from Pioneer. It only did 3.95GB disks, the laser was stuffed after writting 1000 drives and the cost was 21,000GBP or over 40,000USD. Give it a few years and they will be as cheap as chips.
The benefits for write-once media are actually pretty clear. Suppose you've got to keep audit trails for a database containing financial data; writing it to write-once media is a pretty good way of doing it, since it's then easy to show that it wasn't tampered with. Rewritable media is useful for other things (e.g. live data).
First, you only prove that it's not been changed -after- it was written to the medium, which don't bring you much unless you verify the medium after writing it.
And you can do this with read-write media anyway, by writing a *tiny* bit of information on non-changeable media. Put an ad in the NY-time with the SHA-sum of your hard-disc, and you've got pretty good proof 5 years from now that it's been unchanged ever since.
Put an ad in the NY-time with the SHA-sum of your hard-disc, and you've got pretty good proof 5 years from now that it's been unchanged ever since. I don't know about that... Five years is a long time to find a hash collision. So what happens to your strategy when a weakness is announced? Do you tell your auditors that it was good enough five years ago?
Let's put it another way... You give me a SHA1 hash and five years. If the money's right, I'll give you back a dataset that matches that hash wi
However, what ar the odds that the dataset you produce will make sense in the given context? Very high, actually. Presuming I have the original data to provide context, I can fiddle with white space, unallocated disk blocks, executables (since they are not likely to be executed from backup nor examined closely), whatever. Without the original data, then all bets are off. You have to assume an attcker would have access to the data in question.
As pegr pointed out, there cannot be an unbreakable hash and hashes may only buy you time. However as someone else pointed out, if you divulge further information on your file, while it won't make it impossible to have a different file with the same hash, it will make far less likely that someone can bring out a file with the same hash that also has that same length, is a valid bzip2 stream and, after decompression, has the same internal structure (is a valid OOo document). There is a finite number of files
No, unfortunately this is incorrect. bzip2 probably will ignore data after the end of the stream, plus it has future usage blocks that are likely ignored. OOo also has areas that are ignored by the editor. (I'm generalizing, I didn't check either program, but many - can I say most? - programs have such areas.) Using those areas you can add whatever padding data you need to "fix" the hash after adding your fake data.
Recording the file length makes it harder, true. But if you are the one generating the hash yo
You clearly haven't been reading Slashdot this morning [slashdot.org]. In fact it's only a lousy 1000 times better - clearly a rip-off by the optical disk makers to give you less capacity than you thought.
Wasn't the trick behind holographics that a part of the pattern could be used to reconstruct the entire pattern? That would make restoring the data on a dropped (and shattered) holographic cube much more convenient..
"unless they expect thos things to have liftimes in the thousands of years range." Of course, as the drives are likely to be around for about five years and you'll be able to find a servicable used part for a decade after that, a thousand year life time might not serve much purpose.
The trouble with that kind of 'archival media' is that once you realize you need the archive you have nothing with which to read it anyway.
You're better off carrying the data live on some form of redundant array of inexpensive dev
Yes. Do you think that CD-R and CD-RW technologies came out at the same time? CD-R technology was available several years before CD-RWs, so at that time it was CD-R or nothing.
What kind of library has £9000 to spend on a single piece of computer hardware? It'd be substantially cheaper to buy a computer and four of those 1 TB hardisks that were mentioned yesterday, and they'd be rewritable!
Or they could spent the £9000 on, y'know, say... books.
A library isn't always a public lending library. Another type of libarary that could actually might have use for this type of storage solutions (not necessarily exactly this one) is what I would call historical research libraries. Their function is to protect the material and at the same time make it more accessible to people. It's not unusual for these libraries to have a serious digitizing projects so that the originals don't have to be disturbed (especially if they are physically deteriorated). Just the
The type of library that is a copyright library (i.e. receives a copy of every published book) rather than a public library (which is what you are thinking of). Think about e.g. university libraries, the British Library, the Library of Congress, that sort of thing. Obviously a public lending library isn't going to want one of these things, but then you don't go to a public library when you want to find a bit of obscure data.
From the article:
Holographic storage offers extremely fast data transfer rates - currently up to 160Mbit/sec, though there are plans to increase this.
When you have a multi-Terabyte system to backup AND verify within a short window (say 4 hours), speed trumps price just about every time.
What is the cost of NOT having a backup?
ZombieEngineer
Of course, you can build a multiterabyte disk-to-disk backup system with gigabit transferrates out of common of the shelf hardware for less than $1000.
The cost of having backups can certainly be made a lot less than $18000.
With ten years experience working with enterprise class mission critical systems, I've seen those arguments (and those systems) many times. And yet in my experience, the 'rated uptimes' seem to be some definition of 'when the system is up and working the uptime will be 6 9's', because between everything from bugs through randomly incompatible hardware through firmware upgrades through operator (yes, the vendors own certified technicians) error, the actual on-line time for that kind of system usually isnt ev
Holographic storage offers extremely fast data transfer rates - currently up to 160Mbit/sec, though there are plans to increase this...speed trumps price just about every time.
I could be wrong, but are you implying that people will use this because it's got 160Mbit/sec write time? Keep in mind that this is 20MB/sec. That's a little low for the standard harddrive, and you can increase it by adding more drives in a sequential raid. If that's the speed, then it absolutely isn't a good reason to use this.
The only advantage this actually has is information density. One 600GB disc is going to be pretty tiny compared to an array of harddrives designed to get the speed up.
Is that worth it for a library or bank? My inclination would be no. A couple hundred harddrives in a SAN is probably a better idea.
The market will be those individuals that absolutely, positively need the discs to be tiny, and nothing else matters. Because this tech isn't going to do anything else better than what we've already got.
160 Mbit/sec? That's about 33% of the speed of current hard drives, maximum. Remember, transfer speeds are normally measured in MB/sec (or, more accurately, MiB/sec) not Mbit/sec. And for hard drives there are RAID solutions readily available. The things that might be interesting for customers are reliability, power consumption and (kind of) storage space needed. It might be that these things beat hard drives in these sections. Storage capacity, speed and price of hard drives are definately not yet within r
Because magnetic media fails, badly, often, and at any time.
It is in NO way a long term backup solution.
And you don't expect the first generation of this system to fail?! Heh.
Magnetic doesn't fail as much as you make it sound. We have 100s of TB backed up on 400 GB Tivoli tapes and rarely lose a tape. If we do, its not the media itself... a pin from a tape will get stuck in the drive (from the tape being mishandled -- someone dropped it a few times.) The media itself is still usable.
for a high density archival format, but I can't see where this even comes close.
The manufacturer rates it at 50 year archival life, with no specifics about how that number was derived (is that an average? guaranteed for every piece of media? until an error rate of "x" is encountered? under what storage conditions?).
It's a proprietary solution, from a single startup company - what are the odds that a reader is going to exist in 50 years? Note that the manufacturer specifically warns of a lack of backward compatibility when they state "Drive is backward read compatible for three generations; 18-24 months between generations." Having an archive of data which is inaccessible doesn't get you much.
Thanks for finding the information I was just going to lookup on their website. 50 years may not be optimal, but it's a lot better then the only competitors, hard drives or burned DVDs, which usually fail under 10 years. The fact that it's write-once is another plus, since even software bugs can't damage the data. Your other point is valid, but secondary. If your DVDs or HDDs have degraded beyond readability, they're useless no matter how many readers you have. And if the life-span of the reader is longe
Personally, I think the only uses for a 600GB write-only-once drive are backups, a DYI Nuclear Weapons for Rising Countries Kit (or similar content), taking "snapshots" of the Internet, and storing the known digits of pi, largest prime numbers, and other interesting numbers. Then again, there's also the thought about using them for file-servers, and server logs, but seriously, one-writes are not really that attractive given the price tags. Hopefully, the re-writable media/technology will be available within
Wow, a real product. Every time I read about holographic storage, particularly on Slashdot, it's in the same sort of context in which you'd read about quantum computing or Star Trek-style teleportation. Like this: "Scientists at (fill in name of university) have managed to get (name of particle) to (some verb), a first step toward what could one day be practical (quantum computing, space elevators, carbon nanotube frisbees, or whatever). They used a (system you'd never be able to afford) to (do something even your grandkids won't be able to do), and predict that the process will be commercially viable in (about the same amount of time it will take us all to get cold fusion reactors installed in our cars)." Nice to see something like this actually come to market!
I agree with the parent on this. At least it's not vapourware. Always the same debate with new technologies, especially storage - too expensive, something else is better etc. etc. Goes all the way back to floppy disks vs. ethernet. The first hard drives were around 20Mb, and cost a lot more than the 15 or so floppies they replaced.
What would be great is if someone knowledgeable had a look at the technology and made an educated guess as to whether it will be cheap in mass production. I'm pretty sure the f
We need go no further than Slashdot's search engine to nail these bozos: Same guys last year [slashdot.org] making claims that they were shipping that year. Here they are [slashdot.org] the year before along with some other crooks making claims. And earlier in 2005 [slashdot.org]... why do they get so much play for vapor?
Ooh, here's a good one on some guy trying sucker people with funding for his spintronics [slashdot.org] drive that will bring miraculous storage to the masses. He already has pricing worked out!
One potential use I can think of is selling/renting really high definition movies, TV series or collections of movies. For example, 10 seasons of "Friends" in ultra high definition would surely take up a lot of space. For that use a single disc with a huge capacity is perfect.
The disc in question is much more elegant and cool than a stack of bulky, noisy hard disks. Elegant and cool may sound petty, but they sell for certain kinds of people with too much money. They even sell RCA cables for more than $18,000.
Is amusing. It's got the pointless wave abstract graphics I usually see on sites with nothing to say (now, of course, I'm not claiming this, these guys seem serious in general).
Their slogan is "data at the speed of light". Because, they use lasers and holographic technology, do you get it? It's a very smart slogan.
But the reason I'm writing this post is this site reminded me of the International Association of Virtual Reality Technologies (IAVRT) which was supposed to bring Neuronet upon us, and they wamnted to fund this by selling "neuronet domains". They have shut down for a "few weeks" until they hit some major partnerships. Quite some months have passed since.
Check their domain page still with the same message (and notice the uncanny similarities in design with InPhase Technologies):
This product has been "Coming Soon" for a couple of years now. I think this is the third or fourth time this no-product startup has gotten an article posted on Slashdot. It is slow (180Mb/s is in no way "fast), under-capacity (600GB is a waste of time), overpriced, and unproven. If you want near-line storage, use SATA, if you want archival, use tape. I don't see much of a market for this thing.
Its a brand new technology, which means that the initial models have to absorb or get allocated a lot of the development costs, and therefore the price restricts the models for only those who can afford it and have a genuine need for it.
Good thinking (Score:5, Funny)
Good thinking. I mean, if they were launching the disk without the drive (or even the other way round) it would be a lot less likely to succeed.
Re: (Score:3, Insightful)
Re:Good thinking (Score:5, Interesting)
So says you.
Bleeding edge is always a ridiculous expense. The people who are willing to be there already know who they are. That you even raise this question means that you are not.
OTOH, neither am I, but that's not the point. The point is, this is the first commercial volley of a new technology, which means that a few years hence it will be cheaper with even higher data densities.
Meaning, potentially, something like the entire run of every season of every Star Trek series ever... on one disc.
Parent
Re: (Score:3, Interesting)
We've been doing this for a few years actually, as a "roll your own" solution. We currently use removable drive carriers from DataStor, and 500 GB Seagate disks (first ATA, now SATA). We also use foam-padded locking carriers that are take off-site every day. We do ~1.2 TB of backups every nigh
Re: (Score:3, Interesting)
You obviously responded without reading my whole comment (tsk...tsk!).
Alright, so say you buy the $18,000 drive and 200 discs as $180 a piece. You spend a couple of months saving all your highly valuable data, put it in a vault and wait a 30 years. BUT, the next year, the company that created the $18,000 and their proprietary storage goes bell
Re: (Score:2)
Re: (Score:2)
Re:Good thinking (Score:5, Funny)
Good thinking. I mean, if they were launching the disk without the drive (or even the other way round) it would be a lot less likely to succeed.
Yeah, that would be like a game company shipping a console before any games are available for it. Err...wait...
Parent
Re:Good thinking (Score:5, Informative)
Parent
Re: (Score:2)
First, you only prove that it's not been changed -after- it was written to the medium, which don't bring you much unless you verify the medium after writing it.
And you can do this with read-write media anyway, by writing a *tiny* bit of information on non-changeable media. Put an ad in the NY-time with the SHA-sum of your hard-disc, and you've got pretty good proof 5 years from now that it's been unchanged ever since.
Re: (Score:3, Interesting)
I don't know about that... Five years is a long time to find a hash collision. So what happens to your strategy when a weakness is announced? Do you tell your auditors that it was good enough five years ago?
Let's put it another way... You give me a SHA1 hash and five years. If the money's right, I'll give you back a dataset that matches that hash wi
Re: (Score:3, Informative)
Very high, actually. Presuming I have the original data to provide context, I can fiddle with white space, unallocated disk blocks, executables (since they are not likely to be executed from backup nor examined closely), whatever. Without the original data, then all bets are off. You have to assume an attcker would have access to the data in question.
Cryptoanalysis of SHA1 [wikipedia.org] has already weakened it...
Re:Good thinking (Score:4, Funny)
"What is a 'file', granpa?"
Parent
Re: (Score:3, Interesting)
However as someone else pointed out, if you divulge further information on your file, while it won't make it impossible to have a different file with the same hash, it will make far less likely that someone can bring out a file with the same hash that also has that same length, is a valid bzip2 stream and, after decompression, has the same internal structure (is a valid OOo document). There is a finite number of files
Re: (Score:3, Interesting)
Using those areas you can add whatever padding data you need to "fix" the hash after adding your fake data.
Recording the file length makes it harder, true. But if you are the one generating the hash yo
Re:Good thinking (Score:5, Funny)
Parent
Re:Good thinking (Score:5, Informative)
2) Transfer speed
3) (600 gigabytes) / (600 megabytes) = 1 024 times better
Parent
Re:Good thinking (Score:4, Funny)
Parent
Re: (Score:3, Informative)
Re: (Score:3, Interesting)
Re: (Score:2)
Although, at that price, I admit, it seems exorbant, unless they expect thos things to have liftimes in the thousands of years range.
Re: (Score:2)
Of course, as the drives are likely to be around for about five years and you'll be able to find a servicable used part for a decade after that, a thousand year life time might not serve much purpose.
The trouble with that kind of 'archival media' is that once you realize you need the archive you have nothing with which to read it anyway.
You're better off carrying the data live on some form of redundant array of inexpensive dev
Re: (Score:2)
Yes. Do you think that CD-R and CD-RW technologies came out at the same time? CD-R technology was available several years before CD-RWs, so at that time it was CD-R or nothing.
libraries? (Score:5, Insightful)
Or they could spent the £9000 on, y'know, say... books.
Re: (Score:2)
Re: (Score:3, Insightful)
Re: (Score:3, Insightful)
I'll pass (Score:5, Funny)
It is all about data transfer speed... (Score:5, Interesting)
Next time I'll hit the preview button. (Score:2)
What is the cost of NOT having a backup?
ZombieEngineer
Re: (Score:2)
Re: (Score:3, Informative)
Of course, you can build a multiterabyte disk-to-disk backup system with gigabit transferrates out of common of the shelf hardware for less than $1000.
The cost of having backups can certainly be made a lot less than $18000.
Re: (Score:3, Insightful)
Re:It is all about data transfer speed... (Score:4, Informative)
I could be wrong, but are you implying that people will use this because it's got 160Mbit/sec write time? Keep in mind that this is 20MB/sec. That's a little low for the standard harddrive, and you can increase it by adding more drives in a sequential raid.
If that's the speed, then it absolutely isn't a good reason to use this.
The only advantage this actually has is information density. One 600GB disc is going to be pretty tiny compared to an array of harddrives designed to get the speed up.
Is that worth it for a library or bank? My inclination would be no. A couple hundred harddrives in a SAN is probably a better idea.
The market will be those individuals that absolutely, positively need the discs to be tiny, and nothing else matters. Because this tech isn't going to do anything else better than what we've already got.
Parent
Re: (Score:2)
But why? (Score:2)
Re:But why? (Score:5, Interesting)
It is in NO way a long term backup solution.
Parent
Re: (Score:2, Interesting)
It is in NO way a long term backup solution.
And you don't expect the first generation of this system to fail?! Heh.
Magnetic doesn't fail as much as you make it sound. We have 100s of TB backed up on 400 GB Tivoli tapes and rarely lose a tape. If we do, its not the media itself... a pin from a tape will get stuck in the drive (from the tape being mishandled -- someone dropped it a few times.) The media itself is still usable.
BTW...
There is a need... (Score:5, Insightful)
The manufacturer rates it at 50 year archival life, with no specifics about how that number was derived (is that an average? guaranteed for every piece of media? until an error rate of "x" is encountered? under what storage conditions?).
It's a proprietary solution, from a single startup company - what are the odds that a reader is going to exist in 50 years? Note that the manufacturer specifically warns of a lack of backward compatibility when they state "Drive is backward read compatible for three generations; 18-24 months between generations." Having an archive of data which is inaccessible doesn't get you much.
Re: (Score:3, Insightful)
Your other point is valid, but secondary. If your DVDs or HDDs have degraded beyond readability, they're useless no matter how many readers you have. And if the life-span of the reader is longe
Uses? (Score:2)
Then again, there's also the thought about using them for file-servers, and server logs, but seriously, one-writes are not really that attractive given the price tags. Hopefully, the re-writable media/technology will be available within
A real product? (Score:5, Interesting)
Re: (Score:3, Insightful)
Always the same debate with new technologies, especially storage - too expensive, something else is better etc. etc. Goes all the way back to floppy disks vs. ethernet. The first hard drives were around 20Mb, and cost a lot more than the 15 or so floppies they replaced.
What would be great is if someone knowledgeable had a look at the technology and made an educated guess as to whether it will be cheap in mass production. I'm pretty sure the f
Re: (Score:3, Interesting)
Ooh, here's a good one on some guy trying sucker people with funding for his spintronics [slashdot.org] drive that will bring miraculous storage to the masses. He already has pricing worked out!
While I'm sure that sooner or later one of thes
Help me... (Score:5, Funny)
Ultra high definition media (Score:3, Interesting)
The disc in question is much more elegant and cool than a stack of bulky, noisy hard disks. Elegant and cool may sound petty, but they sell for certain kinds of people with too much money. They even sell RCA cables for more than $18,000.
Forget the capacity... (Score:3, Funny)
Their site (Score:3, Interesting)
Their slogan is "data at the speed of light". Because, they use lasers and holographic technology, do you get it? It's a very smart slogan.
But the reason I'm writing this post is this site reminded me of the International Association of Virtual Reality Technologies (IAVRT) which was supposed to bring Neuronet upon us, and they wamnted to fund this by selling "neuronet domains". They have shut down for a "few weeks" until they hit some major partnerships. Quite some months have passed since.
Check their domain page still with the same message (and notice the uncanny similarities in design with InPhase Technologies):
Wavy green lines header [iavrt.org]
Bottom line is, wavy green lines aren't very convincing, we need high res demos of icy cubes storing TB of data, come on!
This is STILL just worthless, and vapor... (Score:4, Interesting)
SirWired
Re: (Score:2, Insightful)