Slashdot Mirror
Shaking Hard Drives Instead of Spinning?
Twyko64 writes "A UK startup called Dataslide aims to develop 'hard drives' made of oscillating sheets of LCD-screen-like material with piezo-electronic actuators and many, many read:write heads. A 'hard drive' could be the same size and shape as an LCD screen. I wrote a this piece on Techworld about it."
I don't know about you lot, but my LCD monitor is an awful lot larger than my hard drive. Surely all those extra heads are going to be really expensive, too?
"I Know You Are But What Am I?"
is there any benifit to this? i mean, it sounds like if we had this to start out with, our current hard drives would be an improvement. just because something is different doesn't mean it's better.
While I appreciate innovation, I think companies should really try to improve the current state of solid state storage devices. Obviously, no moving parts mean fewer points of failure. Also, other than saying that these devices could theoreticly be better than current spinning disks and flash memory, this article is pretty scant on hard specs about the tech. I guess it's way too early for them to release such information, but I'd like to see some specs on it. Like how they are going to cancel out background noise vibrations. Seems to me like this technology would be very exposed to faults due to things like that, perhaps even small vibrations due to loud noise/etc.
stuff
A 'hard drive' could be the same size and shape as an LCD screen
Personally, I prefer my harddrives to be less than 12 inches square... (12x12 = 17" diagonal)
I could see this as possibly useful for a slim computer/tablet sort of thing, but I'd imagine that I could get more oomph out of a slim computer with a 0.25" thick CF card.
This "new drive" seems to have all the disadvantages of a drum, plus another: it doesn't spin. Instead it just shimmies back and forth.
Well, maybe the new magical material will handle this OK. With the old drums, spinning them up often took several minutes because of the huge inertia (weight was often in the hundreds of pounds for the bigger ones... disaster when the bearings seize and the drum smashes through brick walls!)
I think this works by someone with a nothing story putting a link to it here and so people click through and huzzah! Hits come a rolling.
And wow, that is a poorly written article too.
"For lovers of irony we might note that this feature is about shaky technology. But don't knock it. Hummingbirds hover, they hang in mid-air, because of their vibrating wings. The apparently impossible can happen. A violin's shaking strings produce music. "
It was like, shaky...humm, Word Thesaurus, give me shaky words to use and I will use them all in my closing.
The idea seems to be that a vibrating sheet could move, while a grid of read/write heads could stay in place, just so something moves to generate a changing magnetic field. While that's certainly true, a spinning disc could also have mutiple heads per arm, multiple arms per disc, and so on. Getting a closely packed array of read/write heads is an equal challenge in either case, and having the surface move continually in the same direction is much easier than having it oscillate.
This would affect what shapes a drive could be manufactured in, but that's unlikely to matter enough to make the idea catch on.
Who is John Cabal?
The good question is, is vibrating better than spinning? Because we don't have the luxury of nonmoving harddrives at the moment.
I think it could be. You'd need less moving parts, and you'd be able to put them in places that are more easily accessible and replacable.
On the other hand, consider that we do already have such a comparison: power supplies versus harddrives. Other than the fan, power supplies have no moving parts...sort of. But actually the two coils that make up the transformer within a power supply vibrate at 60Hz (in North America), and I've been told that power supply failure normally happens because these wires slowly start to fuse as a result of this stress (culminating in a short)
So which fails more, power supplies or hard drives?
Mod me down and I will become more powerful than you can possibly imagine!
I can just imagine the racket this thing would make. As shake velocity increases to reduce seek time, so will the inertia of the object being moved. Your laptop would take on a life of its own, as it bounces across the desk like a thing posessed.
My rights don't need management.
it has an Antec quiet case and a SATA (that I like to think is quiter than a standard IDE 7200 RPM ChugMaster).
Hmmm let's think about this for a second...
SATA is Serial ATA, a bus format.
Other formats are IDE, E-IDE, etc.
Do SATA drives spin? Sure they do...
Do they spin as fast as non-SATA drives? Sure they do...
What's different on them? The bus...
Does the bus make any noise? No...
So why exactly do you think that SATA matters one way or the other on noise?
Oh and these drives, if they ever become more than a pipe dream, would almost certainly vibrate at ultrasonic frequencies.
I am disrespectful to dirt! Can you see that I am serious?!
First of all, with so few specifics in the article, one is left with speculating. Speculating tells me that a hard drive with a lot of heads is MUCH more expensive than a regular hard drive. The heads, and the mechanisms for controlling them are probably the most expensive part of a hard drive. So I would think and sheet like drive with a whole lot of heads and a mechanism for controlling the sheet is going to be ridiculously expensive.
Of course, they might have a solution for this, but the post, the article, and the company's web sites leave so much unsaid, we may never know. My guess is we'll never see this. There are many other storage technologies that sound signifcantly more promising than this. And solid state still has a long way to go as well, and as a nother poster pointed out, no moving parts... Sorry if I don't leave a post-it note on my monitor about this one.
The guy, in this article, says that flash rom "is faster" than discs. If you're comparing a plain flash eeprom chip (nothing fancy) and the latest hard drives, this is clearly false. My current hard drive sustains a 50-60 Mbytes/s fingers in the nose, and a lot of fancier hard drives can get much more throughput than that. That's about 16 ns/byte. I don't know of a lot of flash eeprom chips that have such a low access time. Of course, you could always "stack" up several of them so as to widen the data path (128, 256, even more maybe? bytes in a row). That would solve the read time. Write time is still pretty much a lot longer than on most hard drives, though. Even if you widened the data path a lot. Which would cause other problems. Just a thought. I think the author was a bit quick here. Motionless storage devices may be the future, but I don't see it coming any time soon. There is still a lot to be doing in solid-state memory before we can achieve this.
Do you mean the 1" LCD on the front of my phone or a 55" LCD TV?
Look, mom, I can store 5KB or my etch-a-sketch!
:wq
Not necessarily. If you've got "many, many" heads, then you have several options. One would be to have some fixed heads above several tracks, eliminating seek time (great for swap space). Another would be to partition the disk into platter groups, with a separate R/W head serving each group. The separate actuator arms could use the same pivot point and magnet assembly. I don't think you'd need a special controller to prevent head crashes, the head assembly only sweeps ~1/4 of the disk surface anyway, adding another arm at the opposite corner wouldn't interfere. It could probably be done within an existing 5.25" form factor, too.
Just junk food for thought...
The article is totally content free, and reads like it was written in german, translated to french, and then translated to english. "I wrote a this piece on Techworld about it." Yeah, I can tell you "wrote a" this piece, pal. Next time cut the crap with butterflies and hummingbirds and tell us how the hell a piezo drive actually WORKS.
If a job's not worth doing, it's not worth doing right.
More often than not, power supplies fail because of the fact that they are the first line of defense against the electrical supply with all its surges and spikes. Those spikes cause damage to capacitors and voltage regulators that builds up over time until the part fails. The result is that the power supply ends up delivering the wrong voltage (usually higher than desired on one rail, lower or zero on another) and often pulsating DC.
I've only had two computer PSUs fail. One of them went open on the output, but both coils of the transformer seemed to check good. (I didn't pull it out of circuit, so I can't be certain, but the resistance seemed reasonable.) The other one shut itself off repeatedly. After analysis, it was hitting a thermal cut-off because the fan had stopped spinning.
I've had many laptop power supplies fail, but that's always a cable break or short. I have had three such supplies replaced and a fourth that just started sparking....
Never a single case of a coil shorting. A coil shorting would just result in a voltage drop if it happened on the secondary or a voltage boost if it happened on the primary. It would take a very serious short before you noticed it, unlike motors where a short often means that the motor won't have enough strength to start.
More than that, the part of a hard drive that fails is almost never the motor. It's usually something stupid like a bearing leaking oil all over the platter or a head sticking somewhere and then either gouging the platter or snapping off and then gouging the platter.
The real question is whether micromotive hard drives would be more reliable than spinning ones. Depends. How are those devices lubricated (or are they lubricated)? What prevents a head crash? I assume that the heads aren't supported by a cushion of air, which would be an improvement, but beyond that, they still have the same potential mechanical issues, only now there's more than one or two heads to deal with. The more heads, the more interconnects, and thus the more potential points of failure.
This sounds an awful lot like probe-based storage. If it is, the advantages are in terms of increased density, not increased reliability. We won't know about reliability until those things are widely deployed. Until then, it's just conjecture.
Just my $0.02.
Check out my sci-fi/humor trilogy at PatriotsBooks.
Heat generation is one problem.
you have to speed up and slow down the point several for every half cycle.
This is a lot of energy, even assuming it is all on 'springs' you will get some mechanical loss due to friction.
This will likely be much more heat and power consumption then a current rotating drive.
What about the cost of many heads? Right now a hard drive is a small number of expensive heads, and a large area of cheap media. This could cause the cost to skyrocket.