Slashdot Mirror
Next Wave Of Hard Drive Tech: Perpendicular Recording
angrytuna writes "New serial technologies are set to replace standard SCSI and ATA (Advanced Technology Attachment) interfaces over the next two years, even as hard-disk drive manufacturers prepare for an entirely new form of bit storage. Perpendicular recording will replace longitudinal recording in storage devices, placing bits on end instead of lying them parallel on the disc surface, thus dramatically increasing the possible storage density."
I am less concerned about the amount of stuff I can put on a hard-drive, and more concerned that the next time I boot up my computer, that stuff will still be there, as harddrives get more and more high-tech, the reliability seems to be taking a big nosedive, how will this effect the reliability of future drives?
In the days of 250GB hard drives, who cares? All I'm concerned about is the speed of drives. Lets improve that for once...
HDD manufacturers said they expect to start replacing 3.5in. disk drives with smaller 2.5in. devices in enterprise products sometime within the next year.
Why would they want to do this? Has it something to do with vibrations (or even shattering a disk) due to the extreme rpm's that these drives are running?I don't know much about this stuff, so could someone please enlighten me?
The best weapon of a dictatorship is secrecy, but the best weapon of a democracy should be the weapon of openness.
Of course, both of these are non-magnetic. And holographic memory is still research-only, as far as I know.
I wonder, will magnetic storage (in any number of dimensions) ever get eclipsed by non-magnetic ones like these?
Nothing is so smiple that it can't get screwed up.
I think RAID is going to come into its own in this environment of the future. Everyone will have an array. The obvious speed benefits of having drives doing tasks in parallel will be hard to ignore with these kind of storage densities.
Perhaps the next big leap will be creating an 'array in a box' which is sold as a single unit. Imagine how many 2.5" HD mechanisms you can fit in a 5.25" disk drive bay, or even 2 3.5" bays. Then imagine them all operating in parallel.
That might solve the problem, or at least make things feasible for a while yet.
HBI's Law: Frequency of calling others Nazis is directly correlated with the likelihood of the accuser being Communist.
...engineers are working with software developers on a way to dramatically reduce power consumption by maximizing the number of 0-bits in memory.
The grain of truth to this joke: There is a well-known technique that reduces the number of 1s in words transmitted on a bus by inverting words that are more than half 1 (and setting an extra bit indicating that the word has been inverted). The idea is to reduce the number of transitions on the bus lines, as a change in state is what dissipates power.
That's assuming current speeds. Well, as data gets more dense, the access speed inherently gets much faster, assuming the RPMs stay constant. If physical size stays the same, random access can't really get too much slower. So what is it that is going to be bad about terabyte disks?
So what? All that means is that it's about time for solid-state storage to come into it's own...
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
those allready exist, and pretty common to. Many companies sell 2 drive raid setups with firewire connections. Also I have seen people put 4 2.5 inch HD's in a raid setup, and inside a Shuttle XPC of all places.
:(
I agree on the raid stuff though. I think soon i'm going to start making all my computers with at least a raid 1 setup and even better a 0+1 setup. HD's are getting cheap. RAID interfaces are getting very common, and SATA seams to be bring RAID with it. 4 WD raptors in a box could make for fun. And noise
Oh, to answer your question then:
Instead of the magnetic field changes being lateral, they are vertical. Don't worry, the substrate is deep enough. It's really just another way to write smaller. Instead of long skinny areas being charged front to back, or back to front, the areas are oriented up and down.
My other car is a 1984 Nark Avenger.
I have seen a few posts from folks not quite understanding how the "bits-on-end" approach works. Some were speculating that it might be holographic, multiple layers, or 3D and such. It is not at all that complicated as they are making it out to be. I heard it best described from Alan Shugart who started the company called Seagate. On an episode from "The Computer Chronicles" back in 1984 he described it as standing the magnetic particles on end to fit more in a given area, which is similar to how a cord of wood could fit into a given smaller area by standing them up on end instead of laying flat. So it really is simpler than you might imagine. Of course the implementation is anything but simple. This is especially evident by the fact that this idea was known as a way to increase storage density back in 1984, when even 200 million bits per square inch was not in a consumer product yet. It was merely in labs with thin film head technology poised to become the next big thing in a short time from that year.g er.php.
By the way, you can see old episodes of "The Computer Chronicles" at the Prelinger Archives collection.
http://www.archive.org/movies/prelin
I believe Slashdot had a story about that a while ago. Good stuff! Great info can be had through those old episodes about computer history.
>>>>>> Chewie, take the professor in the back and plug him into the hyperdrive.
They also talked about drive sizes changes (3.5in -> 2.5in)
Bah, why always smaller???
Current HDDs store 50Gb/in^2, and area increases with the square of the radius. That single inch decrease results in literally half the platter area (not counting the spindle). OTOH, with even current areal densities we could have 1TB 5.25" HDDs. THAT would make me a happy consumer.
But no, that would make too much sense. Instead, they'll shrink the drive, requiring radical new (and untested in the wilds) technologies just to keep up with the same overall size.
Hey, I can appreciate smaller in most aspects of technology. But as long as we store data on spinning platters, where surface area matters, bigger, up to the width of a typical case (ie, 5.25in), makes a WHOLE lot more sense. Hell, use 10" platters and design the case around the HDD lying parallel to the MB for all I care, as long as I have obscene amounts of drive space.
Then again, I probably count as one of the few people who considered the Quantum Bigfoot series a great idea - Large, cheap, somewhat slower drives. For most uses, as long as a drive has a "reasonable" seek time and transfer rate (ie, within an order of magnitude of other modern drives), size matters more than speed. Most of us don't do realtime DV processing, we store tons of what amounts to largely offline content (ie, a huge CD changer would do just as well, other than for the drive we keep our OS on).
You can develop the negative yourself very quickly and cheaply. The slow and difficult part of photo developing is transferring the image from the negative onto paper. But if you are scanning the negative directly onto the computer this isn't necessary - The negatives are a far higher resolution than printed photos in any case.
That's... odd.
;)
When you say uncompressed DV resolution... why use a format that isn't even DV if, by all likelihood, you're using something that came off DV? Is it to preserve the 32bpp? Or is this something you rendered yourself?
Excuse the qusetions, just a curious video n00b... I though working with DVs ~215MB/min was bad enough... less than five minutes of footage per gig! Aaargh! High Density resolution is going to murder hard disks!
Diplomacy is the art of saying "nice doggie" whilst looking for a rock