New Hard Drive Idea

I was contemplating how we might improve the speed of a Hard Drive so that it basically had no wait times, or even a motor, etc. I surmised that we might be able to build a device that uses a laser, in the form of the platter, and what this would do is simple.

As the laser would be identical to the platters size we could just have it read the entire platter itself without needing any heads, then access would be based on a file record and all it would do it read that section of the platter to access the data.

So it would be silent, etc. No moving parts, just a giant laser being used to read the platter. Any thoughts on whether or not this would work?

 

Log in or Sign up to hide all adverts.

It's fine in theory, but you can't read a normal magnetic hard drive platter with a laser, so you would need a new platter made of a new material that stored information some way other than magnetically. You would basically be starting from scratch. What material would you use, and how would you write to it? Also, I assume that you would still at least need a moving mirror or lens or something to control what part of the drive the laser is scanning over.

 

Log in or Sign up to hide all adverts.

What do you mean, "laser, in the form of the platter..."? Up to that point it sounded like you were describing an ordinary CD.

 

Log in or Sign up to hide all adverts.

Hard disks are about to be superceded by flash drives, instant response, silent and no moving parts, they are already available on some upmarket video cams and will appear in computers when they can have enough memory, probably within 3-4 years time.

 

Imagine a hard drives platter, and above that a laser that is as large as the platter. The laser itself isn't focused on a single point, rather there are basically points on it that read from the surface of the platter, but without moving. So you'd have a light dance.

 

I'm so ready for that it's not even funny. They also produce much less heat and consume less power than magnetic hard drives. Thus in turn, laptops will be lighter, because the batteries won't have to be as big, or on the flipside the same size batteries would last even longer.

 

You forget 1 key point:
You have to put a recieving 'eye' lense in their somewhere. Laser beam hits disk, laser bounces off disk, laser either hits the 'eye' or doesnt, eye records bit (and thats an optical CD Drive.)

-Andrew

 

Nope, you can't just "spray" laser light all over a surface and expect to pick up any specific information with some sort of detector. It requires controlled scanning/sensing in order to select what you want.

Just imagine a standard magnetic disk harddrive with a read/write head just as big as the disk itself. There's absolutely NO way to make it select anything specific on the platter. Same principle.

 

Magnetic hard drive technology seems to be keeping ahead of flash technology. A while ago I bought a 250 GB hard drive for around $65 - that same amount of money would only get you around 4GB of flash storage. Yeah, in 3-4 years you might be able to get hundreds of GB of flash storage for that price (although that seems a bit optimistic to me) but by then people will probably be buying hard drives with tens or hundreds of terabytes.

 

I'd rather have this 32GB NAND flash drive laptop than even a 250GB magnetic HD laptop.
If I need more space I could just get an external hard drive. That could be a magnetic one, but for operating system and boot files: flash>>>>>magnetic anyday of the week. To be honest, I'd rather PAY for a flash hard drive than get a magnetic one for free.

http://www.engadget.com/2006/05/23/samsungs-q30-ssd-with-32gb-flash-drive-on-sale-in-june/

...they forgot to mention much less heat output too.
I'd have NO PROBLEM paying extra for the advantages listed above. Less expensive doesn't always mean better.

 

I don't think lasers can work on magnetic platters.

There are 2 moving parts in a hard drive. The heads and the platters. There are 4 heads total on a typical hard disk, 2 heads per platter, top and bottom and there are 2 platters on a typical hard disk. The heads are moved by an arm for tracking.

Please Register or Log in to view the hidden image!

What we want to do here is to improve the performance of the hard disk.

I have some ideas:

1. Instead of using 1 arm, we can design the hard disk to 2 arms, one on each side of the hard disk. This will double the performance.

2. Instead of using 2 arms, we can increase it to 4 arms, one on each qtr of the platter.

3. Don't use any arms at all. We will use "static heads" where there are several tiny heads mounted on a circular platter, each head reading and writing at the same time. The hard disk platter will be sandwiched in between these two circular platters. This way, we will totally eliminate the motors as everything will be static. The hard disk platter doesn't need to revolve anymore as the circular platter which contains hundreds or thousands of tiny heads will read and write directly to the magnetic media.

4. We don't even need the hard disk to be circular. You can even have it at odd shapes like a triangle or a square.

 

If you look the next time you buy a harddrive you'll find mentioned the number of "Cylinders" which is what you've referred to as platters. (Cylinders were the original recording media before records, which is why the stack of platters is named so.) You'll also find a reference to the number of heads, for the most part they will be 16+.

The greater the heads, the higher then density the higher the speed equals a higher capacity for synchronous read-writes.

I state synchronous here because this is what defragmentation deals with. when your files fragment on the harddrive they cause alot of harddrive search time for the fragmented pieces when a file is executed. This causes a greater wear and tear on the arms and longer execution time in general. (The "disc" or cylinders continue to spin at their normal speed, the heads hunt for which sectors the information resides.)

A little anecdote in regards to the nature of the harddrive verses a DVD-rom.
A friend of mine had a large collection of images for their shop that he decided he wanted to place in storage, the problem was however it would take a number of DVD's to store the images in full printsize, so this required the images to be first placed into a compressed file then placed upon the DVD.

This worked well, until it was time for him to repopulate the computer with the library, every time he tried to extract the files from the compressed file on the DVD it would fail. Even attempting to move the compressed file from the DVD would fail.

I had the fun job of fixing the problem, it was a simple task I had to result to copying an ISO of his whole DVD and then using a Drive emulator to access the DVD image. I was able to extract the images directly with no fuss from the image and even copy the compressed file to the drive.

The problem was that the compression system required to have a number of heads available to deal with the compression algorythm which would have seemed like a fragmented file, The DVD drive could only deal with a small range and while it to seeked back and forth for the fragmentation it would cause itself to error. The Harddrive with it's multiple heads was able to deal with the seek problem and in turn deal with the error.

(Moral to the anecdote, either don't compress to a file then shift to a disc or just go out and by an external harddrive for extra storage.)

 

Just for the record, the physical platters still exist, of course.

Please Register or Log in to view the hidden image!

A cylinder is nothing more than a vertical slice through ALL the platters that is one sector in width. (Picture a narrow part of a slice of pie stacked on several others.)

The original media was NEVER cylindrical in shape. The original hard disk drive was name by the Burroughs Company in the late 1970s. It had a single platter mounted vertically and 48 fixed heads - 24 on each side and mounted in the clamshell cover. The entire thing was hermetically sealed and belt driven by an external 1/4 hp motor. It was mounted in a steel frame made of angle iron and the whole assembly weighed just over 100 pounds. It took two full-grown men to change out a drive.

Please Register or Log in to view the hidden image!

It's total storage capacity was a whopping 1 KB.

Please Register or Log in to view the hidden image!

I had the good fortune of working on with a group of the first batch produced ( at $4,800 each) - serial numbers 9 through 18. The technology worked so well that it's been disc-shaped platters all along.

 

Would a sphere work better in storing large amounts of data?

 

I kind of meant like this http://en.wikipedia.org/wiki/Phonograph.

 

One word---

Spintronics.

http://en.wikipedia.org/wiki/Spintronics

 

Why can't the whole platter be read by the laser instantly? Is that not possible? Then there'd be nothing to access.

 

darksidZz, I think I got what you mean.

We need to use several tiny lasers to do the job. Normal CDROM drives have only 1 laser moved by a tracking motor : (http://electronics.howstuffworks.com/cd6.htm).

What we can do here is to do away with the tracking motor and have a fixed arm instead. The fix arm will contain several tiny lasers which covers the whole platter across.

I don't know whether this works though. Might be too complicated.

 

Please explain specifically how the laser would read the platter. Suppose you shine a laser on the platter and some of the light is reflected, some absorbed, and some passes through. What exactly are you measuring in that situation, how are you measuring it, and how does it turn into data?

 

The problem with producing immobile optical readers is that would need a laser and receiving lense as wide as the data track is wide all allong the radius of the disk. At a data track of 0.5 microns with 1.5microns seperating two track would make 2microns the maximum width of each laser and lense pair. given theres a 6 cm radius of disk (minus the center), this would amount to a huge number of pairs, and almost surely be more costly than flash memory of equal size (and would be less efficiant in terms of power consumption, and heat). Secondly, having all those tiny lasers and eyes means the drive will be extremly fragile; if even one of those lasers becomes unalinged, your drive wont be able to read data (it basically toast.)

-Andrew