IBM Reinvents Punch Cards

grim_thing writes "I.B.M. scientists say they have created a data-storage technology that can store the equivalent of 200 CD-ROM's on a surface the size of a postage stamp. Writing in the current issue of the journal IEEE Transactions on Nanotechnology, researchers at I.B.M.'s laboratories in Zurich report that they have achieved a storage density of one trillion bits of data per square inch, about 25 times as great as current hard disks." Reuters also has a story.

Comments

Perhaps not a disk-replacement, but....

[ajs]

So, that would be 120Gb in the size of a postage stamp. Not bad. Even if it takes a long time to write and longer to read back, this could wipe out tape archival for most backup purposes!

Re:Perhaps not a disk-replacement, but....

[Andre060]

So, that would be 120Gb in the size of a postage stamp. Not bad. Even if it takes a long time to write and longer to read back, this could wipe out tape archival for most backup purposes!

I'm not so sure. One of the most important factors in backups (if not the most important) is Cost per megabyte of storage. The article does not talk about cost.. How easy will these devices be to mass produce? What will they cost?

Andre060

Re:Perhaps not a disk-replacement, but....

[mobets]

Maybe it will just become the New Tape (tm). Just immagine the capacity of a whole tape cartridge of the stuff.

Re:Interesting

[leuk_he]

They just sold their Hard disk unit to hitachi. And a few days later they report a new storage format.

Makes you think...

plausible cause for dumping HD business?

[f00zbll]

Since there isn't a whole lot of details about this technology and exactly when it will show up in store shelves, it's kind of hard to guess IBM's plans for this technology. How plausible is it that IBM has something totally knew to replace HD technology and this is just another related development. Whether this can/could/should/would replace HD is hard to say without real data, but it might provide a clue. IBM might have some other bleeding edge technology lined up for mass storage, which lead to the development of this product?

Where are all the inovations?

[peterdaly]

I seems every couple of months one of these "new storage breakthrough" comes along. What happens to them? Where do they go? It seems like these things have yet to make it into consumer (or even "professional") technology. Have heard a lot about high density solid state storage, and stuff like that, yet I still have a platter spinning at 7200RPM next to my feet. Arn't we a little outdated by other technology standards using spinning pieces of metal to store our information, with no end in sight?

These things are cool, but they become science breakthroughs, not news for nerds...stuff that matters? Do breakthroughs like this really matter to us? I am asking this because I really don't know. Where have semi-recent "Breakthroughs" like this made it into consumer technology that you and I can buy today? Or next year?

-Pete

reliability of millipede?

[bbc22405]

So, there's a thousand red-hot pokers, melting a trillion holes in a square of plexiglass. Each poker will make a half billion holes just filling up the chip the first time. Eeek! I assume these chips will come with plenty of error detection/correction, so that if one of the pokers quits, the remaining ones will give you the clues to what was in that 0.1% that you just lost.

But certainly, there must be some sort of failure rate for each poker, and the chip... Is it too soon to know/guess these numbers?

Re:reliability of millipede?

[Drizzten]

From the article:

The Millipede chip consists of a layer of plexiglass a couple of billionths of an inch thick laid on a silicon chip. To write a bit of data, a microscope tip, heated to 750 degrees Fahrenheit, softens the plexiglass and dents it.

To read data, the tip is heated to 570 degrees -- not hot enough to deform the plexiglass -- and pulled across the surface. When it falls into a dent, the tip cools because more surface area is in contact with the cooler plexiglass. That temperature drop reduces its electrical resistance, which can be easily measured.

To erase data, a hot tip is passed over the dent, causing it to pop up.

Those are pretty high temperatures. Might they affect reliability in the long run? Also, what about the actual lifetime of the plexiglass material they're heating up? How many times can it take the read/write/read/write process? Still, it's a nifty idea. Kudos to IBM.

Re:One potential security flaw. . .

[sphealey]

What if one's data contains dimpled chads? How will those bits be counted?

The funny thing is when I am done voting I always turn the card over and check for chads. After the Florida thing I mentioned this to my spouse, who gave me one of those "only someone like you would think of doing that". I finally realized it was a result of the many hours spent in front of an IBM 029 keypunch, followed by 4 hours waiting for the card deck to come back from the machine room. When one hanging chad can kill a day's work, you tend to check for such things. But I imagine the percentage of people with that experience is getting lower by the day.

sPh

Re:possible use...

[TheCrunch]

"maybe we will all have on embeded in us and it wil contain our DNA table"

DNA embedded in our bodies?! What a novel idea.

But on topic, this development is cool. I look forward to having significant amounts of solid-state (well.. less moving parts) storage. It'll open the door for countless new computer applications. Digital voice recorders for example; ~90 days worth of audio in your pocket is very impressive.

How does this compare (density-wise) to holographic storage?

Re:Interesting

[clunis]

Point taken, but you may want to take a quick refresher course in IBM history and corporate culture. To say "the right hand doesn't know what the left hand is doing" does not even begin to capture it. Something like "the right hand doesn't know what the giant squid banana fish love Neptune" would be more like it.

Seriously, big blue is huge. I'd say there's little chance that the folks who shut down the hard drive division knew (or cared) that some branch of the research division were about to make an announcment like this.

Chances are they don't even know now.

80 columns? Surprise! NOT!!!

[alumshubby]

Back around '79 or so, I remember hearing a COBOL trainer (in a corporate setting) assert that in the next century, there would be a language called COBOL, even if there was not way of knowing what it would evolve (or maybe the word is mutate) into. By now, I feel pretty secure in seconding his notion that COBOL, the Legacy Language from Heck, is never gonna fade away. (In fact, as a career option I'm weighing COBOL as a language to concentrate on.) IBM apparently feels the same way, so it's not too surprising that they'd come up with a whole new way to archive all those billions of lines of code in the handy, familiar 80x25 format.

Re:numbers Re:Interesting

[markmoss]

You are right that my numbers (10 years development, 99% chance of failure) are wild guesses - based on 35 years of watching promising new technologies move, or usually not, from the labs. Lots of unexpected problems turn up in trying to commercialize new technology, so most new developments die without ever being produced. If it gets past that hurdle, it will still die in the market unless it is much better than existing technologies, which haven't been standing still while all the problems are worked out. For example, bubble memory once sounded this good relative to competing technology (almost as fast as semiconductor RAM, nonvolatile, and might have been cheaper than disk drives), but by the time it was actually in mass production, semiconductor RAM was much faster and cheaper, while hard drives had shrunk from the size of washing machines to small enough for PC's, became cheaper and more capacious than bubble and not too much slower. There weren't enough applications where bubble was definitely better to support efficient mass production, so it was soon priced right out of the market.

Judging from this report, they haven't taken the first steps to commercializing the hole memory. They are writing and reading with a scanning electron microscope - a lab instrument that probably costs six figures. And they are writing 1,000 times slower than a modern hard drive. It would be nice to have a full backup of the server farm fit on a credit card, but not if it takes days to complete the backup... They need to move it to a purpose-built machine, solve the speed problem, get the cost way down, standardize formats, and get the drives and cartridges on the market. Sounds like ten years - if it is possible to solve the cost and the speed problems in the same machine. Using 1,000 heads instead of one would solve the speed problem _today_, but it sure doesn't help the cost issue. And by the time they are ready to market it, what will they be competing against?

Data Security

[theblacksun]

It's always been a problem with the magnetic techniques of hard drives to actually erase data. Even after formatting and overwriting, you can still read old bits because of the inprecision in the writing techniques (see this article). Would this eliminate such a threat?

Re:numbers Re:Interesting

[afidel]

Judging from this report, they haven't taken the first steps to commercializing the hole memory. They are writing and reading with a scanning electron microscope - a lab instrument that probably costs six figures. And they are writing 1,000 times slower than a modern hard drive. It would be nice to have a full backup of the server farm fit on a credit card, but not if it takes days to complete the backup... They need to move it to a purpose-built machine, solve the speed problem, get the cost way down, standardize formats, and get the drives and cartridges on the market. Sounds like ten years - if it is possible to solve the cost and the speed problems in the same machine. Using 1,000 heads instead of one would solve the speed problem _today_, but it sure doesn't help the cost issue. And by the time they are ready to market it, what will they be competing against?

The core of the Millipede project is a two-dimensional array of v-shaped silicon cantilevers that are 0.5 micrometers thick and 70 micrometers long. At the end of each cantilever is a downward-pointing tip less than 2 micrometers long. The current experimental setup contains a 3 mm by 3 mm array of 1,024 (32 x32) cantilevers, which are created by silicon surface micromachining. A sophisticated design ensures accurate leveling of the tip array with respect to the storage medium and dampens vibrations and external impulses. Time-multiplexed electronics, similar to that used in DRAM chips, address each tip individually for parallel operation. Electromagnetic actuation precisely moves the storage medium beneath the array in both the x- and y-directions, enabling each tip to read and write within its own storage field of 100 micrometers on a side. The short distances to be covered help ensure low power consumption.

While current data rates of individual tips are limited to the kilobits-per-second range, which amounts to a few megabits for an entire array, faster electronics will allow the levers to be operated at considerably higher rates. Initial nanomechanical experiments done at IBM's Almaden Research Center showed that individual tips could support data rates as high as 1 - 2 megabits per second.

Sounds like they are not reading with a TEM but with real devices, the speed problems have already been adressed, as well as power considerations. The only thing left is cost =)