New Storage Technology Idea.

[sao123]

I was thinking about the possibilities of future storage technology. The following is one such theory on how it could be accomplished,
thoughts please.


The actual storage medium would be a 3D Cylindrical Crystal. the crystal would have to be composed of a material that would have at least 2 or more physical states that would be unique, stable, and controllable for the purposes of storing data.
For example one such method could be defined by the pattern of bonds within a unit of the crystal. (Number and orientation)
Depending on the structure of the crystal the number of distinct physical states could determine the amount of information that can be stored in a unit. 2 states - binary item / unit, 8 states - 1 byte / unit, 32 - 1 (32 bit) word / unit.

The interface for the medium, would be read and written via 2 lasers, similar to 3D CDROM Technology. One laser would be fixed pointing at a 0 degree radius at the central axis but be moveable along the Z axis, which the crystal medium would spin about it.
The other laser would be under the crystal cylinder and be able to freely move along the 0 degree radius pointing in the Z axis.
data could be read or written at the intersection of the lasers.
Essentially this system would use a Theta, R and Z cylindical coordinate system to write and retrieve data.
I believe this technology could maximize the amount of storable data in a medium, while able to maintain a high speed transfer rate.

Strong points
1)high speed read/write
2)Less Fragging potential.
3)solid crystal allows for better recovery of internal data recovery even in the event of surface scratching.

Weak points
1)What crystal meets design (physical and chemical) specifications?

What would be some good features or limitations of this design?

 

[Shalmanese]

Weak point: It is still fundamentally mechanical in nature and, thus, failure prone. This would be good for using as anarchival device but probably not the ideal choice for everyday use.

 

[zephyrprime]

How exactly would the two laser read the data? Why wouldn't they end up reading all the adjacent data to?

 

[sao123]

The data would be read at the intersection of the two lasers.
It would take both lasers to determine the status of the crystal unit cell.

 

[zephyrprime]

I understand that it intersects, but how that that do anything? The lasers have to pass through all the adjacent cells and that mucks up any attempt to read the data.

 

[imgod2u]

I agree. If the laser passes through an entire line of data points, how is it that it'll be able to draw the distinction between which data point it crossed with the other laser? You can only read the data that is carried with the laser either being reflected or passing through the crystal. Even if you could somehow adjust the 2 lasers to interfere with eachother when they meet, how would you determine how far (one-dimentional wise) the lasers interfered as.

 

[ScottMac]

Maybe it'd be easier to use the medium to generate an interference pattern, with an optical sensor to read the interference pattern ...which would resemble an array of "black & white" (on/off) dts/stripes/bars (depending on the nature of the material and the interference pattern it generated. Changing the nature of the media would change the nature of the interference pattern (kind of like holographic storage).

FWIW/.02

Scott

 

[sao123]

Actually I was thinking in a more chemistry way.
The entire media crystal as a whole is in a stable low energy state.
A Crystal unit being energized (higher energy excited state) by one laser would make it stand out, and then be readable by the other laser.

 

[Ilmater]

OK, so you explains how its read, now explain how it writes....

 

[imgod2u]

The point is not what energizes and what not, but how do you determine at which point exactly you're reading and writing at. Remember when you shoot a laser at high energy, it'll affect all points in the crystal. A method to do this would be to have each "layer" in the crystal respond to a different wavelength of light and have the laser be capable of various wavelengths of light but that is extremely expensive to make. Sure it saves space, but the added complexity and chance for error is too great.

 

[ScottMac]

"Tunable" LASERs are a fairly recent accomplishment, and I'm not sure they've been perfected yet. A common (desired) use is in network trunking for use with Dense Wave Division Multiplexing (DWDM).

The nature of LASERs makes them inherently a single wavelength (lamda) device during emission.

FWIW

Scott

 

[sao123]

These lasers are perpendicular (they cross at a 90 degree angle.) Laser 1 will energize an entire column of data. Then laser two will only see and read 1 row of energized crystal. In other words Laser 2 (the actual reading laser, cant read un energized crystal).

I forgot to explain writing earlier.
Writing is accomplished similar to reading. There will have to be a third laser opposite the reading laser to write. So it becomes a 3 laser system. The write laser will be specially designed to break, create and orient bonds within the crystal.

See diagram
( I || )
Crystal ( 2^^W||R )
( I || )
<-e------------->

W & R move up and down along the axis.
e moves horizontal. E energized Column I, W reads row ^. W can only see/modify the pinpointed location(2).

 

[sao123]

Oh, I see white space doesnt work. Re diagram.

(.......I,....... ||....................)
(.......I,....... ||....................)
(.......2^^W||R..................)
(.......I........ ||....................)
(.......I........ ||....................)
<-----e------------------------>

 

[zephyrprime]

I understand whay you're saying. It's a sort of row/column reader. Then the only problem is finding a suitable medium.

 

[GoodRevrnd]

Even if you had a laser that could emit at different frequencies wouldn't it still defract when passing through other layers and hit multiple points or shoot off in the wrong direction?

 

[sao123]

while its true that a crystal would refract, reflect and bend normal light,
a laser is so focused tight beam that is passes through the crystal nearly unaffected, and it affects only a localized path through the crystal structure, which give the column effect.

 

[MrDudeMan]

which then enables the reading laser to have a very limited place to get the data you are looking for...makes sense

i still dont understand how it will write to the crystal medium and what kind of information allocation table would it have? how would the lasers know where to navigate and what would move them?


this kind of movement would probably be much slower than a hard drive. you would physically have to move the lasers inches at a time

 

[sao123]

Well the writing would use a different laser to configure(change) the bonds into a specific pattern and orientation, thus writing.
The unit could use similar to the scheme used by RWCDs. Just iImagine a stack of CDs. Some bits detemine the platter, track, and position. This address would be translated into the laser coordinate system, thus giving a position for the 2 lasers..

 

[aka1nas]

2 lasers would not be enough to accurately retrieve the data. I think you would need at least 3 to pinpoint the location to read from(they used 6 in stargate to find the coordinates of a planet, FWIW).

 

[AlexKN1]

my head hurts

 

[Cashmoney995]

Good Idea....but isnt storage technology way up there anyways? I highly doubt this could be a plausible alternative to what we already have. Its over my friend, we already have the ultimate storage device. Its 1 cd that holds 1 Terabyte, thats 1000 Gigabytes. About 20 of these CD's could hold the internet (snapshot). Possibly your method may have some use, but its highly unlikely that it would be a mass method of storage.

 

[sao123]

Yes my friend, we have exceeded my expectations for storage capacity.
However, storage is still almost the slowest unit of the PC in terms of data transfer(throughput).
HDD > CDROM > Removable Disk (ZIP FLOPPPY) > Mouse > Keybd.

ATA 133 UDMA mode 6 is decent.
SCSI is fast, but too expensive, there has to be a faster way.

Admittedly the problem is a bottleneck everywhere digital converts to analog to be stored ina mechanical way.
We need a non-mechanical digital mass storage system.

 

[Souka]

IBM had some crystal tech idea going. I remember reading on-line about it.


Sorry, but no more info.....

 

[jbond04]

This has nothing to do with sao123's crystal storage idea (which is actually pretty interesting), but it does have to do with accelerating mass storage...

What about some sort of RAM based solution? With such advanced manufacturing processes today, couldn't we create a low speed RAM that was cool running and could store massive amounts of data? I guess the only problem would be with maintaining the data after power was cut, but is this a feasible solution?

 

[Howard]

Those are solid-state hard drives.