Increasing SSD Capacity With the Help of Sound
"We’re near the peak of what we can do with the technology we now use for magnetic storage," said Pallavi Dhagat, an associate professor in the OSU School of Electrical Engineering and Computer Science. "There’s always a need for approaches that could store even more information in a smaller space, cost less and use less power." Dhagat said that sound waves could improve storage density in SSD devices.
"This technology should allow us to marry the benefits of solid state electronics with magnetic recording, and create non-volatile memory systems that store more data in less space, using less power," said Albrecht Jander, also an associate professor of electrical engineering working on the research.
There was no information on a working system using this technology, but hopefully it won't be long.
Here's a close-up of this new, experimental technology:
http://www.thatericalper.com/wp-content/uploads/2013/01/Close-up-of-a-record-stylus-on-the-grooves-of-a-vinyl-record.jpg
Sometimes, youhave to edit links back in. The link "Read the comments on the forums" between the article and the comments section lets you edit your posts. You can give it a try from there.
SSD's work by having millions of little "buckets" to hold electrons and the copper wires are a delivery system.
How would a SOUND version of this work?
There's really no way to have a main source that focuses sound at the scale they imply. It's also an added cost.
Maybe I'm dead wrong, but I know how Hard Drives and SSD's work and can't even remotely picture how this would work.
Run for your life if they use Justin Beiber...
Needs more metal, maybe Iron Maiden or something.
I'm sure the case will be sealed and soundproof.
and possibly this:
It isn't clear how this quote relates to stretching or bending small metal elements using sound waves, though, nor how it relates to increasing storage density in an SSD. SSD != SSE. Instead, perhaps he's talking about a way to eliminate the need to move a physical magnetic read head across a disk? Just not clear.
Yes you are.
This article doesn't really make a lot of sense. They want to delay heat assisted recording in HDDs by making SSDs better? Those technologies are about different devices. Why would increasing SSD capacity make manufacturers stop increasing HDD capacity? Unless they are talking about SSDs overtaking HDDs in terms of capacity (which doesn't seem likely here) in which case people would pretty much stop using HDDs.
The speed you would be talking about is the length of the waveform, which exceeding 40kHz is extremely small. In a small controlled environment you wouldn't have to worry so much about propagation loss. If anything I think it's interesting that they are adding a mechanical element to solid state.
It isn't any more clear than the excerpt that Wolfgang published. It doesn't look like they're talking about SSDs in the commercially-available sense, but rather about a new technology that's kind of like magnetic recording ala HDDs, but somehow uses ultrasound to store or help store the bits. So a little like HAMR, but then they go on about a solid state memory device:
The article doesn't say who "researchers" refers to, or how such a device would work. It certainly doesn't look like a HAMR replacement unless they're talking about just the ultrasound transducer itself and not the entire storage device. And it doesn't look like it has ANYTHING to do with traditional SSDs. To get high storage densities, it has to be talking about manipulating bits on a uniform surface at a microscopic scale. So I don't know how they think this can be done solid state without an actuator arm, but who knows. Maybe they're really smart guys or something.
Now let's do some quick math:
If present-day commercial HDD technology supports streaming writes at up around ~120 MBps from a 3-platter device, then that's 40 MBps per platter, or 320 Mbps per write head. This means that each bit on the platter is being written for no longer than 1 / (320 million) = ~3.1 ns (question: are write heads single-bit, or parallel?), and likely much shorter than that. So these guys are either saying that they want to use ultrasound to warp the disk surface in just 3.1 nanoseconds, or else that the write data rate is going to be much slower.
I need a picture. I just don't see how this tech is going to fly.