Toshiba Claims Bit-Patterned Drive Breakthrough 151
CWmike writes "Toshiba will detail a breakthrough in data storage later Wednesday that it says paves the way for hard drives with vastly higher capacity than today, reports Martyn WIlliams. The breakthrough has been made in the research of bit-patterned media, a magnetic storage technology that is being developed for future hard disk drives. Bit-patterned media breaks up the recording surface into numerous magnetic bits, each consisting of a few magnetic grains. Under a microscope, the magnetic bits look like thousands of tiny spheres crammed next to each another. Data is stored on these magnetic bits: One magnetic bit can hold one bit of data. Prototypes of the media have been made before but Toshiba says its engineers have, for the first time, succeeded in producing a media sample in which the magnetic bits are organized into a pattern of rows."
But can they afford it? (Score:2)
I've heard that patterned media will be too expensive to ever mass produce profitably so the industry will probably use HAMR instead.
Re: (Score:2)
Who did you hear this from, a guy in a bar? Give us some details man, inquiring minds want to know. Wikipedia says Seagate is talking about a combination of patterned media and HAMR, but both technologies appear to be years into the future.
I read TFA (Score:5, Informative)
Re: (Score:2, Troll)
Actually, the article actually only claims about a 4x increase (actually only 3.62):
Re: (Score:3, Informative)
uh... 2.5tb/541gb = 4.62 (technically rounded up to 5)
Learn to use the calculator, dude.
Re: (Score:2)
What video do we get this time? (Score:3, Funny)
I'm kind of curious; after the "Get Perpendicular! [youtube.com]" video, how's Toshiba going to top Hitachi in the "silly video explaining your new technology" race?
After reading TFA, I'm almost scared that it'll involve some sort of cartoon magnetic grain orgy.
Re: (Score:2)
Re: (Score:2, Funny)
At least it's not a breakthrough in Heat Assisted Magnetic Recording. I can just imagine a 'HAMR Time' video.
how this differs (Score:2, Informative)
here is a link which might explain things more clearly
http://www.bentham.org/nanotec/samples/nanotec1-1/Piramanayagam.pdf
Re: (Score:2)
100TB, here we come! (Score:3, Funny)
Quick explanation (Score:5, Informative)
In current drives a bunch of rather randomly sized and shaped magnetic grains are basically "glued" to the surface of the drive, and the collective orientation of a certain number of those grains (called a domain) determines whether you've got a 1 or a 0.
In this, instead of dumping grains onto the surface, they're using lithography to carve very precise grains onto the disk, which can be made much smaller and more identical in shape, than the random ones allowing for vastly higher storage densities. It's basically applying the same technology used to make computer chips to make hard drives. The technology has actually existed for a while, but the cost per bit to pattern lithograph a hard drive has always been huge; I guess Toshiba has figured out how to bring it under control. Cool stuff.
Re: (Score:2)
Some people will probably need a car analogy to sum it all up.
Re: (Score:3, Interesting)
Existing disc platters are like parking in a field. Each car has to find its way to a spot that is clear of surrounding vehicles, and there is no pre-defined organization of the parking spots. So typically, extra space will be wasted in pathways for cars to get in and out, and there will be the inevitable mishaps with cars trapped in their spots or with no escape.
The new method precisely defines each parking spot, and there is an optimal amount of space provided for every car to
Re: (Score:2)
I don't know about a car, but 2.5 Tb/in^2 makes the tracks one 645th the width of a human hair.
Re: (Score:2)
I'll take a crack at that.
Point to point distance of: car vs train.
A car is like existing drive technology. Depending on who's driving (manufacturer), the road (materials used), the point-to-point distance (inverse of capacity) of a trip varies. If the driver isn't an expert, he might be all over the road (more magnetic surface used) - adding distance to the trip (decreasing capacity). Additionally, if the road is rough (randomly sized magnetic grains) distance is added to the trip.
A train, however, util
Re: (Score:2)
I want to see the PizzaAnalogyGuy [slashdot.org] version!
Re: (Score:2)
Damn, I'd even settle for the BadAnalogyGuy. Where did he go?
Re:Quick explanation (Score:5, Funny)
His last post ended with:
I've had it. I'm switching all my machines to Linux today.
See ya'll in 24 hours.
That was May 11th..... I guess the transition didn't go so well.
Bit-patterned pepperoni pizza (Score:2)
I want to see the PizzaAnalogyGuy [slashdot.org] version!
In his absence, I'll come up with one. Say you're storing data on a pizza in the layout of "sausage" (seasoned ground beef pieces) over the pizza's surface. If you just put the sausage on top of the cheese, you need fairly large areas of sausage to make discernable areas. But if you lay down a grid of pepperoni first, you can put a piece of sausage or nothing on each piece of pepperoni.
Re: (Score:2)
Cairo vs. L.A. (Score:3, Funny)
Some people will probably need a car analogy to sum it all up.
Ok - in LA you have seventeen or so lanes of traffic. But because they are all headed the same direction, they all sit at a standstill pointed the direction they are supposed to be going.
Now compare that to Cairo, which has cars going every which way along with camels and a million pedestrians per square mile. In Cairo everyone sits at a standstill, but they may not be pointed where they want to go, with the single side benefit that they can bu
Re: (Score:2)
Replace "figs" with "oranges" or "melons" or "flowers", and you could still be describing Los Angeles. Minus the camels.
Mostly.
Thanks, firehose (Score:4, Insightful)
You know, I deliberately posted a different version of this summary [slashdot.org] specifically because the summary that was selected here is a lazy cut-and-paste of the poorly written lead of TFA itself.
And not only wasn't my superior summary not selected, but it's been deleted from the firehose page [slashdot.org], where it should appear between Minority Report Style Iris Scanners in Mexico [slashdot.org] and Cats Lies and the Research PR Machine [slashdot.org].
Slashdot has gone from valuable to random, and is going from random to stupid.
Re: (Score:2)
specifically because the summary that was selected here is a lazy cut-and-paste of the poorly written lead of TFA itself.
But that was specifically why sampenzus picked this version. He is all about stupid. Have you not seen the rest of the shlock he posts?
Re: (Score:2)
You think that /. employees are actually using any sort of qualitative decisionmaking criteria when selecting articles for the main feeds?
That ended long ago.
Now they take whatever has the highest +/- ratio when the bell rings to churn the ad stream.
I'm not sure who deleted my version of the link or why, but I'm sure that it involved a long, heartfelt, gut-wrenching decision to do the right thing. Not.
Re: (Score:2)
Slashdot has gone from valuable to random, and is going from random to stupid.
I'm guessing by this comment and your high UID that you are new around here?
Re: (Score:2)
Re:Thanks, firehose (Score:4, Interesting)
It indeed might be the reason that this got picked.
See - with this copy-pasted summary - there is much less chance of it being wrong - and thereby lesser chance of ridicule.
Any issues in the summary/article - the buck can be passed to the article in question - again the editor escapes censure.
This way, the editor does not need to think too much about the article, rather a non-thinking way of copy-paste can produce the maximum results with minimum effort and minimum pay for the editors.
A more cynical view could be that with a perfect summary, people reading the article will be lesser - thereby decreasing the ad revenue for the articles - even though I do not fully subscribe to it - as per Hanlon's razor [wikipedia.org]
I read your summary - it is a perfect summary - it summarizes the main points of the article properly and in an ideal world - all summaries should be written that way. /. where quality is given scant recognition.
But, the fact it was not picked seems shows the sad state of affairs in
I see a pattern here... (Score:2)
Today: New technology overcomes previous limitations!
Tomorrow: Limit of technology predicted! Oh noes!
Day after: New technology overcomes previous limitations!
Re: (Score:3, Funny)
So how is this any different than existing HDDs?
This is a hard drive on speed, known as ADHD (Advanced Digital Hard Drive).
Re: (Score:1)
Re:I'm not a hardware guy (Score:5, Funny)
Yeah, but there will be a million regular HDDs for sale that are mislabeled as ADHDs...
Re: (Score:2)
Telephone modems operate at the hardware level too, and yet store 8 bits per transmitted symbol.
It should be possible to do the same with the magnetic symbols on a disk, if the head could read the "level" of the magnetism (from 0 to 255) at each location. Even if we could only do levels 0 to 3, that would allow us to encode 2 bits per spot on the drive instead of just 1 bit.
Re: (Score:1)
I'm guessing existing HDDs aren't bit-pattered.
Re: (Score:2, Informative)
From what I read in the article, it looks like Toshiba's reduced the number of magnetic grains per bit from a few hundred down to just a few. Otherwise it appears everything is the same.
Very, very small isolated domains (Score:5, Informative)
Re: (Score:2)
Re: (Score:3, Informative)
I think you said exactly the opposite of what you meant to say.
If the boundaries between the bits are more "discreet", then they are more hidden.
If the boundaries between the bits are more "discrete", then they are more distinct, and presumably will interfere with each other less often.
-Your friendly neighborhood Grammar
Raaiiiiiiaaaaaaiiiin (Score:2)
If the boundaries between the bits are more "discreet", then they are more hidden. If the boundaries between the bits are more "discrete", then they are more distinct, and presumably will interfere with each other less often.
-Your friendly neighborhood Grammar Nazi
For someone nitpicking someone else's lingual mistakes, it's ironic you missed the fact that it *wasn't* an example of incorrect grammar!
:-P
BTW, *this* [lileks.com] is a grammar Nazi.
Re: (Score:2)
To continue the picking and pedantry...
Nobody said it was, so now the weight of all that irony now rests squarely on your shoulders.
Re: (Score:2)
Secondly, the OP's implication that this was a grammar issue was clear. Holding them to that isn't pedantry- if they meant it as a catch-all for language issues, it shows that they don't understand or care about the difference, which doesn't say much for their authority on the subject.
Re: (Score:3, Informative)
He said DISCREET, not DISCRETE.
Re: (Score:2)
>> His comment made sense to me; discrete as in separated.
>
> He said DISCREET, not DISCRETE.
I'm sorry but you've failed the Turning Test for this week. Please try again next week.
Re: (Score:3, Informative)
I "called him out" because he was saying almost exactly THE OPPOSITE of what he meant to say. I wasn't picking on a simple typo or common misusage ("it's" when "its" is proper), but a time when the wrong word choice could cause confusion for the reader.
Re: (Score:2, Funny)
You're not being very discrete in the way you're handling this.
Re: (Score:2)
So, are you saying that not only will the capacity increase but the reliability will increase as well?
Re: (Score:3, Insightful)
RTFA, it's enlightening, not all that technical, and not TLTR. And you really should learn how the hardware works; writing software is a LOT easier if you understand the underlying mechanics.
Re: (Score:2, Informative)
Bit-patterned media breaks up the recording surface into numerous magnetic bits, each consisting of a few magnetic grains. Under a microscope, the magnetic bits look like thousands of tiny spheres crammed next to each another. Data is stored on these magnetic bits: One magnetic bit can hold one bit of data.
Just like every other hard drive! Hooray for the future!
"Toshiba says its engineers have, for the first time, succeeded in producing a media sample in which the magnetic bits are organized into a pattern of rows."
Just like every other hard drive! Oh, wait...
Re: (Score:3, Interesting)
I'm really exited to hear this:
Toshiba expects the first drives based on bit-patterned media to hit the market around 2013.
When was the last time we heard about a new tech breakthrough that wasn't followed up with "5 to 10 years" ...Though it might be 5 years by the time the price drops enough for the avg consumer.
Re: (Score:2)
The first hard drive using a read head based on giant magneto-resistance was commercially released about 10 years after the effect was discovered.
Re: (Score:3, Insightful)
So basically, they reinvented the hard-sectored disk? *confused*
Re: (Score:3, Funny)
Yes, but this one has sectors one bit long.
Re: (Score:2)
No, that would be magnetized areas read in circular columns.
Re: (Score:3, Informative)
Except that current HDDs use a wider area of surface to write the data too as compared to this.
Re: (Score:3, Funny)
What's your problem? I'm not upset. Not even the slightest little bit.
Re: (Score:2)
It's bad enough we already can't tell whether a "megabyte" is binary or decimal. Now we can't tell whether a "bit" is physical or virtual.
Re: (Score:2)
psssst
See how I used the word bit there to describe something really tiny? Maybe that has something to do with its referencing the physical "bit". As in, it overlaps both the digital bit and the physical bit. /whisper
Re: (Score:2)
I'm not upset. Not even the slightest little binary digit.
Here, fixed that for ya! Now should be totally and unambiguously clear for tech nerds too, you linguistic oriented geek!
<mumble>(could never understand how the humanity still manage to exist, with such fuzzy and ambiguous [wikipedia.org] ways of communicating and mentally operating with notions and notion and terms that do have a clear definition)</mumble>
Re: (Score:2, Troll)
It's bad enough we already can't tell whether a "megabyte" is binary or decimal. Now we can't tell whether a "bit" is physical or virtual.
A megabyte always has been and always will be binary-based.
MB is not an SI scalar, nor did it ever pretend to be, nor is it conflicting with the SI scalar M.
The only confusion comes about when you try to insist that MB is infringing on some sacred, arbitrarily-based notion that all major scalars must be factors of 1000.
The "classical" units and scalars are themselves ambiguous. What does M mean? Meter? Mass? Minute? Mega? Milli? What does G mean? Gram? Giga? The gravitational constant? Is K kilo?
Re: (Score:2)
You don't think they've bit off more than they can chew with a bit in their mouth?
Whether they call it "bit" or "mite" is rather irrelevant, IMHO, as long as it doesn't lead to another stupid acronym. What's important is that it isn't ambiguous, and it doesn't seem to be.
Re: (Score:2)
If there's a 1:1 correlation between "number of physical bits" and "number of virtual bits"... does it really matter? "Megabyte" being binary or decimal matters, because they're different sizes with the same name - a "binary" megabyte (1024^2 bytes) and a "decimal" megabyte (10^6 bytes) hold different amounts of data.
A "physical" bit on this storage, and a "virtual" bit in memory hold the same amount of data, from what the article says... so why would this cause all kinds of confusion? One is the physical
Re: (Score:2, Informative)
For the uniformed: with today's technology, a 1:1 correlation between data bits and magnetic "bits" is nearly impossible. We have to interleave data bits with clock bits, so we are able to count runs of equal bits. So the data bits are encoded on this interleaved stream of data and clock/sync bits before it is actually stored in the physical medium. If the bit-patterned layout doubles as a clock/sync mechanism we can store only the data bits (with error correcting codes too, of course).
Re: (Score:2)
Interesting to know - but as a mostly uninformed sort... is this an argument for using some sort of other term for them, or does this suggest that there is NOT a 1:1 correlation between physical bits and virtual bits?
I'm not clear on what you're trying to say here, other than to share this info about how it works.
Re: (Score:2, Interesting)
There is not a 1:1 correlation, but there might be now. With all physical bits being data bits we could gain up to 100% more data bits on the same area.
Re: (Score:2)
Re: (Score:2)
Parent has history backwards. Disks were invented, and measured in megabytes, back when bytes were not necessarily 8 bits and computers were not necessarily sold as "binary" machines. The typical disk record was 80 bytes long, since it came from a Hollerith card. The IBM 1401 [ibm-1401.info] was typically sold with 4K bytes of main memory. Four thousand 6-bit bytes.
What does the fact that computer cards ... (Score:2)
were originally the same size as the US currency in use in the late nineteenth century have to do with anything?
Punchdcards were later shifted to 96 columns and the dimensions of the card shrunk from the old format (which I still remember fondly along with my old IBM 29 keypunch,) to these tiny punch cards.
None of this made any sense back in the nineteen-seventies and none of this makes any sense now.
A typical record was an integer divisable fraction of the 19k 3330 DASD write buffer length when I was worki
Re: (Score:3)
I'm sure the next thing will be the bits (in base 10 of course) that are available before the ECC, clock bits, sector relocation table, and other niceties are put in. Similar to raw capacity versus formatted capacity, except before the critical HDD functions are factored in.
Re: (Score:2)
Of course if someone hands you a computer I'd suggest you place money on a byte being 8bits. Chances are it is. Also, I'm not seriously suggesting we start using KiO, MiO, GiO, etc... Just poking a little fun at the industry.
Re: (Score:2)
We'll call them phybibits and vibibits and then you won't have to worry.
Re: (Score:2)
Because HDDs aren't the past and aren't going away anytime soon? It's no different than the fact that 3.5" floppies and tape drives and tapes are still sold despite being proclaimed as being "the past" and dead.
Re: (Score:2)
Doesn't mean shit.. its still dead tech, just like the fucking floppy that you think isnt dead.
Re: (Score:2)
Yes [frys.com].
Re: (Score:3, Insightful)
Because solid state's main hold back has always been capacity. Magnetic media capitalizing on it's main selling point isn't unexpected.
Besides, I see the future is being a mix. Solid state for my boot drive containing all my programs and such. Magnetic media for my Bittorrent and iTunes drive where I need space but not speed (afterall write speeds to those drives are limited by my dirt slow internet speed, and read speads only have to be quick enough to keep up with playback).
Re: (Score:2)
Re: (Score:2)
> Once SSDs get up to around 2TB, people won't care about how large hard drives are.
Yes, and 640k is all that anyone will ever need.
Meanwhile, there are already those of us that not only imagine a use for more than 2TB
of disk space are actually using much more than that already. Just let granny loose with
a hi def video camera and watch that disk space quickly disappear.
Already still cameras seem like something to inspire an Odo rant from DS9.
Re: (Score:2)
But today not only are we storing movies, we are storing them encoded at state of the art resolutions in the native format that they are available at commercially. We can increment up with twin-stream stereo video (aka 3D), and maybe a few doubling of resolution. There just isnt any orders-of-magnitude storage demand increases on the movie storage front.
Previously, there was always some next exponential thing th
Swapping as usual, I see (Score:2)
My write speed is usually limited by the fact that some large memory program gets swapped out to disk
If you swap for a reason other than putting /tmp on a RAM disk [wikipedia.org], then you need more RAM. If you cannot fit more RAM in your PC, then you need a motherboard that isn't nearly a decade old like the one in my PC.
GarageBand and iMovie drive (Score:2)
I see the future is being a mix. Solid state for my boot drive containing all my programs and such. Magnetic media for my Bittorrent and iTunes drive where I need space but not speed (afterall write speeds to those drives are limited by my dirt slow internet speed, and read speads only have to be quick enough to keep up with playback).
Then what kind of storage do you recommend for a GarageBand and iMovie drive? These are for the interactive creation of works, so they need speed, but the works are digital signals as opposed to text, so they also need space.
Re: (Score:2)
Because solid state's main hold back has always been capacity.
SSDs have plenty of capacity. The problem is price per unit capacity. You can get 1 TB SSDs . . . they just cost over $3000 [newegg.com]. (But one of them claims 1.4 GB/s read/write speeds. Nice.)
Re: (Score:2)
I make $N million a year selling hard drives. But wait! Flash drives are the future! I don't need to spend any money making my hard drives better; I'll just sit out the last 5-10 years' worth of profits in that business. Someone else can have them; I don't mind. Really. No worries at all. It's only money, after all.
Re: (Score:3, Insightful)
Re: (Score:2)
However the smart hard drive vendor would realize that spinning platters are headed out the door
Like 3.5" floppies and tapes, right? Oh wait...
Re: (Score:2)
Re: (Score:2)
but if the only thing you sell are spinning disk hard drives, like Western Digital and Seagate,
Oh [newegg.com] Rly? [seagate.com]
Re: (Score:2)
Re: (Score:2)
Spinny disks just upped the ante here by a factor of 5.
Meanwhile, serious SSD's might become cheap enough for a consumer to consider eventually...
Knowing that something like "entire movies stored on your computer" is inevitable someday is one thing. Expecting it tomorrow is another.
Sometimes it takes 10 or 20 years for reality to catch up.
Re:Advancing the Past (Score:4, Funny)
Obsolete 2TB spinning platter device = $99.99
New hotness 640GB flash device = $14,500.00
Oh, well then... There's nothing really stopping me from being the next Governor of California [msn.com]. Jump on the bandwagon for my inevitable victory!
Re: (Score:2)
With SSDs prices and lower capacities, HDD still have room to grow a little bit. If you can get a 3.5" HDD with whooping 10TB of space for 300usd or whatever they price it at, HDD will still be a viable choice for a long time, specially for desktop PCs.
Re:Advancing the Past (Score:4, Insightful)
Spinning platter HDD are not going anywhere until SSD prices become CHEAPER per byte than regular HDD. Regular HDD technology is still improving, as witnessed by this article, and there is no telling when it will slow/stop, or when SSD technology will slow/stop. It is perfectly conceivable for technology to get to a point where SSD can no longer increase in capacity without increasing physical size, while spinning platter may continue to increase capacity with the same form factor. It is also possible that spinning platters may some day greatly exceed the performance of SSD.
You sound like the kind of person who would be surprised to hear that tapes are still widely used for backing up and archiving data.
Re: (Score:2)
SSD needs to become cheaper per byte, yes. But not to the extreme of becoming cheaper then magnetic media in order for it to take off. At some point, it will become "cheap enough" to become useful to a wider range of customers.
Right now, we're a bit above $2/GB for MLC-based SSD while magnetic drives are in the 0.07/GB to 0.12/GB range (3.5" 1TB $70 or 2.5" 500GB $60). Which is about a 16x cost diff
Re: (Score:2)
In the beginning, SSDs were ridiculously expensive because it was for early adopters and others with plenty money to spend. The prices were in freefall and people were making extreme extrapolations based on that. But then the bottom hit in that this many flash chips makes a rather expensive bill of materials. I bought my SSD in April last year. The same model is still on sale for 83% of the cost and the cheapest SSD of same size is now down to 65% of that cost.
At those rates it will be many, many years unti
I'm sure I'll get tore up on the details... (Score:2)
But, bubble memory was more expensive than the hard drives they were intended to replace. Now, we are focused on using various flash memory schemes to accomplish the same feat. Is flash memory related to bubble memory? Who knows, but it fills the same niche, so I'm saying that flash enherited bubble's legacy, to replace hard drives with solid state, non volatile memory
As far as Optical Buses go, isn't that pretty much dominated by Fibre Channel? We use it to connect processors to processors and SANs to proc
Re: (Score:3, Funny)
Personally, I avoid dropping electronics, especially ones with moving parts, on to the floor. Or any other surface, for that matter. Helps a lot.
HTH, HAND
Re: (Score:3, Insightful)
Backing up a 1TB hard drive is a trivial concern when measured against the cost of a 1TB SSD.
The gulf in price between spinny disks and SSD buys a lot of redundancy.
Re: (Score:2)
actually, you can still buy the 2.88MB floppy drive, IBM MF356F-815MB, but it's $130. The SCSI TEAC FD-235J 5604 2.88MB SCSI floppy can be had for $290
Re: (Score:2)
Get offa my lawn you damned kids!
Just fyi, but this was not due to a 'unique' floppy drive format, but rather a defective floppy drive controller.
The device was sold as being 1.44MB PC compatible, but the floppy drive was unreadable by any other 'standard' 1.44 MB floppy drive.
http://boards.straightdope.com/sdmb/archive/index.php/t-10293.html [straightdope.com]
This might seem trivial to you, but in the days before USB flash drives, it was a major pain in the ass. Toshiba could have avoided the whole thing by just licensing a
Re: (Score:3, Funny)
Dude, if you need to see an iPhone to know if someone is gay you are in serious trouble! Have you noticed how your ass keeps being fondled every time you are in a crowd?
Hrm, I don't seem to have that problem. Maybe I should buy new pants.