Slashdot Mirror
300 gigabytes in the size of a DVD?
Rollie Hawk writes "Although storage space is no longer the premium it once was, physical backups and external media have been slow to catch up. While recordable DVDs may be fine for backing up a single workstation, large servers are still forced to rely on swappable drives and tape backups. But holographic disc technology could be changing all of that in the very near future. Holographic Versatile Discs (HVDs) have been in the works for some time now by various companies, including InPhase Technologies (formerly part of Lucent) and Japan's Optware (which claimed to have made the first recording of a movie on a holographic disc last year). InPhase's HVDs, scheduled for release in 2006, are said to hold 300GB of data, 60 times that of a conventional DVD with only a slight increase in size. That translates to more than a day's worth of HD-quality video. Not to mention the drives themselves can read and write at ten times the speed a normal DVD drive. One of InPhase's partners in HVD research, Maxell, is working towards even more storage on a 1.6TB disc."
Turner Network Television recently aired a commercial off of the InPhase Tapestry drive. Maxell built that drive for InPhase.
Data backup has become very expensive for some of my customers. The amount of data a company of even minimal size (50 employees) goes through in a day blows my mind. We've been investing every option but none are cost effective (except when a hard drive goes).
My dilemma is that as backup storage (such as the HVD) gets bigger, it seems that hard drives quickly outpace the new form of backup storage. 1.6TB discs sound great, yet I'm weary of having that much data on an easy to break/burn/steal disc. 300GB is more feasible as I can see making a few copies of the backup "just in case."
Nonetheless, the write speeds listed don't seem all that great, and what interfaces will let us copy data at those speeds? Moving 1.6TG of data off of a server without slowing down user access (24 hours per day with offshore employees) sounds like it will still take hours and hours to back up (if not longer). A recovery stage would take even longer.
For now, I'm happiest with redundancy backups. I don't like mirroring or RAIDx/y or clusters (too many nightmares over the 15 years I've worked with all of it), but having a server dupe itself daily has given us the best turnover and safety margins we've seen, as well as being very cost effective compared to use-once media or (shudder) tapes.
This stuff has been a year or two away as long as I can remember. Someone wake me up when product actually ships...
TODO: Something witty here...
It's not just that it's a fast drive. It goes well beyond the current method of spinning the disc faster and(or) putting the data closer together to increase performance.
"Unlike other technologies, that record one data bit at a time, holography allows a million bits of data to be written and read in parallel with a single flash of light," says Liz Murphy, of InPhase Technologies. "This enables transfer rates significantly higher than current optical storage devices."
That's pretty wild for a single "head" drive. I wonder if this could translate into devices similar to hard drives using similar methods. Hard disks are what I feel is holding back system performance. It's almost always the biggest bottleneck in a system, and has been more or less at a platoe for years, mainly because magnetic media can only do so much in a serial manor.
-Pete
Soccer Goal Plans
Hmm, maybe now they can put the entire Lord of the Rings Triliogy on one disc. now, if you want to put in the extra features, that is a different story...
"The article notes that the transfer rate is at an average of 1 gigabit/second. That is equal to 0.125 gigabytes/second, or 128 megabytes/second, which is a large leap over earlier storage mediums, whose transfer rates are generally measured in Kilobytes/second. In comparison, a 56x CD-ROM drive transfers at up to 8.4 Megabytes/second, and 16x-speed DVDs transfer at 22 Megabytes/second."
That is impressive indeed. But I have a question regarding the random errors etc due to statistical variation. How much resources do you have to devote for error correction (eg parity bit etc) ? And wouldn't it be very power consuming to do error correction at such a high data transfer rate ?
You can't fit a 13cm disc drive into a standard enclosure! Who do they think they're going to sell these to!
"Who is the Journal of Quantum Physics going to believe?" --Stephen Hawking
the Phantom will contain one of these when it comes out...
At my job we use LTO drives( $ 10k per 24 tape drive) + tapes (400 gb compressed on one tape,$50 per tape ) which is ridiculously expensive but works ok . But for home it is not acceptable solution .Right Now I am in desperate need for a backup solution for my home machine - I have 750 Gb and plan go over 1 tb in next quarter. And I am basicaly either have to go with RAID 5 ( ~$1k for 1 TB ) which I dont like since I want incrementally buy more storage - not pay upfront big bucks only to find out that half a year later cost would be halfed. Something liek 300 Gb dvd would be a god sent to me.
A "conventional DVD" has two layers, and holds roughly 9Gb of data. 300 / 9 = 33 1/3, not 60. Even recordable DVDs are well and truly available in 9Gb formats by now, and have been for some time. OK, the media is more expensive than single layer discs, but the technology is in people's PCs now. And how much are these discs going to cost for the first few years of their existence?
Is 300 GB necessary? From a content producer standpoint, I don't want to be able to fit that much content on a single disc... then I can't charge as much for the special 4-disc pre-Christmas release edition.
From a consumer standpoint, I don't need this either, unless I want to archive all my files, in which case it's easier (and cheaper) to have a second hard drive.
I understand there is demand for high-volume storage solutions, but I can't see a mass market for them...
What I do see being very, very useful is the speed upgrade for r/w -- especially for gaming, but I'm sure this applies to other areas as well.
IMO, though, I don't see a big enough demand for this to become profitable for quite a long time -- especially if Bluray or HD-DVD is 'good enough' for the average user.
"Trolls they were, but filled with the evil will of their master: a fell race..." -- J.R.R. Tolkien on Olog-hai
Plenty of people already do error correction at line speed on gigabit communications links with low power costs. To someone developing coding schemes, storage devices can just be modeled as another communications channel.
retrorocket.o not found, launch anyway?
Once the MPAA, RIAA and every other cultural cartel gets a hold of this, it will die like the DAT tape.
They should just release it as a means of backing up data and then figure out the copy protection.
-We get a new storage medium.
-They squable for 5 years.
-Then *MAYBE* they come out with a larger capacity disks with DRM for TVs/movies
Obama's legacy: (N)othing (S)ecure (A)nywhere and (T)error (S)imulation (A)dministration
Even if it is a small change in physical size for the media I'm not too hot on that. I like how our DVD storage is the same size as the previous generation of CDs. The result is that all of my data recorded on CD-R from 8 years ago is still readable and I use it from time to time on our new drives... You can't say that for many other optical or magnetic storage media of odd sizes. Zip Disks, SuperDisks, Jaz Drives... Maybe they're going for a different market, but you think these folks would leave their options open.
There are also good things to be said about leaving the past behind, and not keeping the same physical form factor.
Opinions?
No, they're missing the real advantage to holographic storage: better data integrity. When a CD gets scratched, you lose actual bits. But with a holograph, every little piece of the disk contains the entire image. Try it with a real holograph some time... cover up the bottom half, and you can still see the entire thing. Even break it in two, and you have two identical (albeit dimmer and with a limited perspective) versions of the original. Just so long as they are smart about the way they implement it, and the sensors are a good deal more sensitive than they need to be, this kind of media should be a lot more robust than current CDs and DVDs, and maybe even hard drives.
If you could angle it at around 15degrees you could fit it in. Not sure how that would affect the centripetal force as the disc spins though.
Cool, now I can lose all of my data by just misplacing a single disk. Ain't that grand?
"I'm never quite so stupid as when I'm being smart" (Linus van Pelt)
I could imagine that those things would be great for doing backups, but: Will they be reliable ?
When writable DVDs and DVD burners got affordable I was thrilled at first: Finally being able to backup several GB of data to one not too expensive disk instead of on a stack of CD-Rs ! But then reality hit: Compatibility problems between individual brands of burners and brands of media, quality problems with media, even worse durability than CD-Rs; altogether more or less a total gamble if you want to do backups with that stuff. Now my DVD burners collect dust or are mostly used as CD burners only. So what is a high capacity medium good for if it is not reliable besides making expensive coasters and wall clocks ?
In 1982, I had an Atari 400 (5MHz 8bit 6502) with a tape drive which cost $500. I upgraded its 16KB RAM to 48KB (replacement) for $500, and the tape to an 88KB (double sided) floppy for $500. Now I can get a P4/3.0GHz for $300, a $104 300GB HD, and 1GB RAM for $60. That's 1440x the CPU bandwidth, 16.4Mx the storage, 10.4Mx the memory for a dollar - which is itself worth less than half its value (in noncomputer goods) of a quarter-century ago. And the HD is 1/10th the size (volume), while the other components are about the same size. So it's clear that storage technology has advanced the most during the "PC revolution", by a factor of a thousandfold. The only competing tech is the transformation of my $500 300bps modem and $50:month Compuserve account to a $50 6Mbps DOCSIS modem at $50:month, which is 20-200Kx cheaper for WAN.
I'm all for putting that 300GB into a cheap, tiny device. All the other cheap, even mobile networked computing has created mainstream demand for archive, beyond memory and storage. But I'm betting on it not because storage tech is somehow lagging. I'm betting on it because that industry is by far the highest performing personal computing innovation we've got.
--
make install -not war
Why not? Old floppies are 13.33cm in diameter.
I'm a little confused by some of the responses I'm seeing here. It seems most people think that this new technology would be a bad thing? The largest complaint I've seen so far is: "What if I lose the disk?" I don't know about you guys, but for me this isn't much different then anything else in the world. If your backup is important... don't lose it!
If needed, I have some handy solutions to solve the "How do I not lose my disk" problem.
1. Put it some place that you can remember.
2. If it's super important, make two backups.
3. Tie a string around your finger to remind you to always remember where you put your disk.
4. Ask somebody more responsible than yourself to watch over it for you.
And if none of these work...
5. Buy a small cable. Run the cable through the hole in the center of the disk. Buy a small padlock. Padlock the cable around a large object. Make multiple copies of the padlock key and tape them in various places.
Obviously this is not practical, but it is about as practical as thinking that a new technology is bad because you might misplace it.
http://p2pnet.net/story/7124
Once again, the post-office will become the king of high latency high bandwidth. Hollywood should quake in their boots over this, not on-line file sharing.
If these things are inexpensive enough ond can imagine peer-to-peer postal networks popping up. Say you record half of something on on DVD, and you send it to someone. They send you back half of something, and then you send the other half and so-on. tit for tat.
The problem with the above concept is that it requires the sender and the receiver to actually haveing something each other actually wanted to exchange. But if the disks get big enough you could easily put many things on them increasing the probability that one or more things on their will be something someone else wants to share. It costs you no extra postage to send 1 thing as 100 things now.
So this might blow that wide open. And sharing 100 to 1000 movies per 32 cent stamp, or sharing every single top 40 song for the last 100 years on a single Disk and it wont take long before everyone has every song and movie.
Some drink at the fountain of knowledge. Others just gargle.
For data backup purposes, 300GB would be great. I find myself burning off about 5 DVD's a week worth of data and I can't keep up.
However, once we start thinking about the new kinds of technologies for video distribution, therein lies the problem. For now, say you can put 4 episodes of a television show on 1 DVD. So now, we have 7 DVD's for one television season, plus 9 seasons. Movie studios will not give up that business model. Each of those 4 episodes sells for $35-$50. What happens when all of the sudden you can put the entire series on one HVD or HD-DVD or Blue-ray disc? It won't happen. They will make "higher quality" versions of the same media and still find some way to put 4 episodes on one disk.
What I really want is to spend $50 on an entire television series, with 1 - 5 discs. That is all.
5.25" = 13.34 cm.
Why wouldn't you be able to fit a disc having a diameter of 13cm in a 5.25" enclosure ?
You can't fit a 13cm disc drive into a standard enclosure! Who do they think they're going to sell these to!
Laser disk diehards!
The capacity and speed issues are wonderful. But how stable is the medium? Magnetic media was good for five years minimum. DVDs are good for about twenty before delamination issues threaten the data. In truth there is no stable medium for data. Paper has a better longevity. The best longevity is still clay tablets, but not very practical for the volume of data we need to secure.
Having dealt with data retention for a good 20 years now, I am concerned that whenever there is news of a breakthrough in storage media, it is all about capacity and speed and nothing is said about stability and longevity.
It's great to know how fast the car will go and how much you can haul, but when will the wheels fall off?
How much critical data from our era will acutally survive? I know, how much of it SHOULD survive? That is a different issue, one also faced by data archivists. Let's leave that one alone for now. The burning issue for me is, if I WANT it to survive 200 years, is there a storage medium I can use to insure that? Or do I just make sure there is a clean printout that I seal in a zip lock bag. ( almost joking here...)
Even if it fits in current enclosures, it will never be a consumer-grade product like DVD-RW is now. Home DVD-RW for home theater just dropped below $100, and internal drives for desktops are under $30. Blu-Ray/HD-DVD could get there in 5 years (doubtful, as HD will remain a premium item for most of that time).
Until there is a demand for prerecorded media with 300GB on it, there won't be the impetus to make these items cheap. They'll remain in the computer-room-only expensive category.
What might make this technology fly is not a 300GB, 13cm platter, but a PSP UCD-sized disk for portable media with, day, 20GB on it. However, I suspect that falling flash memory prices will overtake this too quickly for it to have much impact on portable media players, camcorders, etc.
It will be valuable and marketable to the server room customers, but don't expect Dell to include these babies in a $399 desktop for at least 6 years.
Design for Use, not Construction!
In holographic media, read and write operations are usually done using different laser wavelengths. You use a "recording" wavelength to record an interference pattern in the media, and a "reading" wavelength which will diffract into the interference pattern and restore the original image.
These wavelengths need to be different because holographic materials work like photographic films. If you try to read the hologram with a wavelength to which the holographic material is sensitive, you will destroy the interference pattern, and therefore the data.
Wikipedia states that a 532nm laser is used for both reading and writing operations. That means they use a different way to store the hologram. Would anyone have more information about this ?
Yeah, that's "Funny"
Hey, I'm just your average shit and piss factory.
I'm sorry, but this sounds more than a little problematic. At ten times the write speed of a normal DVD, it sounds like buffer underruns would be the rule rather than the exception, unless you were able to use a medium-sized hard drive as a buffer. That means that you're looking not so much at the problem of disc errors from the drive's own hardware, but rather from disc errors because existing hardware can't keep up with it. It would be nice if write speeds could be decreased for compatibility, but this topic hasn't seem to come up yet AFAIK. ((I am not an expert; I'm just saying what I've read.) CD and DVD burners have this ability, but if HVD's depend on such a parallelized method for writing data, then who knows?)
And yes, I agree with the person who complained the loudest at the fact that the discs are larger. (I'll let you argue among yourselves over which of you that was.) After you invent a disc that can hold that much information, making the concession of perhaps even 20% capacity per disc so that the drive fits in existing computer cases is an excellent tradeoff. An important thing about all the other well-known breakthroughs in computer hardware is that it didn't require everything else to radically conform to the new standard it offered. If I want to build a state-of-the-art box with one of these drives in it, I should not have to make major allowances for this device like I would for...say, the motherboard and processor (32 or 64 bit?) or the graphics card (PCI-Express? Maybe also an AGP port for a legacy card?).
If the drive has to be larger, then so does the case. Consequently, you have to have specially-resized versions of existing CD and DVD drives, and other optical devices that were originally designed for another size. (At the very least, you would need an adapter kit to make a smaller device fit properly inside a larger bay.) Only a manufacturer with arrogance that rivals that of Microsoft would require perfectly-good standards to be rewritten just for them.
So, while I would be one of the first to welcome this technology, I still think it requires a bit more thought before the public should get their hands on it.
Not only that, the 5.25" refers to the size of an old floppy disk - not the size of the enclosure.
... and then they built the supercollider.
The storage media for digital theater projectors.
So instead of having to install a bank of hard drives just for single 120-minute movie in uncompressed digital format, you can reduce it all to a single HVD disc plus protective caddy weighing at most 5-6 ounces. This could drastically cut the cost of digital theater projection, since all you need is a small player connected by a high-data rate cable to the digital projector itself. You also have the major advantage of drastically reducing media duplication and shipping costs, too.
the enclosure is 14.6 cm wide.
This is moronic. This is annoying. I am beside myself with frustration. (Well, not really. --I don't actually care.)
The fundamental truth of the matter is that the technology which is readily possible, and the technology which is actually made available to the public, are decades apart. After all. . , why nip the spirit of profit in the bud when you can produce and sell entire production runs of stone-age computer tech one incrementally advanced stage after another? Heck, this keeps the economy 'healthy' during peace times, ensures jobs and an appetite for more and more junk technology. "Planned Obsolescence" is reality.
When everybody gets all excited about the big "new" thing, I groan. We're being led on and sold crap because there are miles and miles of money to be made between now and when the really good stuff is released, which of course, only happens when it doesn't matter anymore.
So who cares? Just let me have enough technology to do what I need to do. Those needs were well met about five years ago, so honestly, I don't really care about any new so-called 'advances'.
And the band plays on. . .
-FL