'Racetrack' Memory Could Replace Hard Drives?

Galactic_grub writes "An experimental new type of memory that uses nanosecond pulses of electric current to push magnetic regions along a wire could dramatically boost the capacity, speed and reliability of storage devices. Magnetic domains are moved along a wire by pulses of polarized current, and their location is read by fixed sensors arranged along the wire. Previous experiments have been disappointing, but now researchers have found that super-fast pulses of electricity prevent the domains from being obstructed by imperfections in the crystal."

Sounds like...

[Anonymous Coward]

...they've updated coil memory.

I've seen it in fibre before...

[simm1701]

I remember reading some research a couple of years ago that somethign similar was done using 100km of optical fibre and a router programmed to keep sending the same stuff around the loop, or it could read it/write it as it came around.

In some ways being slower is definitely an advantage, even with 100km at 10Gb/s you don't have much storage when the bits are moving at the speed of light.

Re:Sounds like...

[AltGrendel]

I think you mean core memory [wikipedia.org].

No, core memory...

[msauve]

stores 1 bit per "core." The article is about a form of memory which continually cycles multiple bits stored as magnetic regions through a single physical ring. The OP is correct in that this is similar to cycling photons through an optical ring.

Looking back, this is all very similar to shift register memory, one of the earliest forms of solid state memory.

Re:I've seen it in fibre before...

[Anonymous Coward]

Mercury delay lines were the cause of a bizarre
computer architecture. The normal form of instructions
had an "address of next instruction" field.

After getting the program to "work", i.e get the correct
answer, the "optimization" stage consisted of working out how
long each instruction would take, and then positioning the "logically next"
instruction at the location just about to appear out of the delay line.

There was no advantage to inner loops that were faster than the
delay round the mercury loop. Unless you could unroll and fit two
repetitions into one trip round.

Of course, all of this was done by hand.

That's not bubble memory...

[Yoozer]

It's Shigawire! [wikipedia.org]

This will bring us one step closer to the Dune Universe. I call dibs on the first load of Spice!

Re:Anything

[dargaud]

I understand the need for air to keep the head flying off the surface of the platter. What I don't understand is the need to have a hole to exchange the air with the outside. Can't they just fill it with neutral gas at the optimum pressure and seal the damn thing ? I say that because I've used hard drive at high altitude and they FAIL often. I mean, if they can do it with salad, why can't they do it to HDs ?

Re:Anything

[Dogtanian]

That reminds me of something I'd almost forgotten; the first Amstrad PC clones (*) that my Dad had at work required you to run a utility to "manually" park the heads on the hard drive before you powered down. Or maybe I'm remembering it wrong.

(*) Amstrad is a British company who (amongst other things) sold the first *really* successful PC clones on the UK market.

Wow, Acustic Delay Memory on the nano-scale

[Glomek]

Am I the only one reminded of Acoustic Delay Line Memory [wikipedia.org] by this?

Re:Plus one addressing

[A nonymous Coward]

I wish! I understand Mel *exactly*. Squeezing the last bit of performance and efficiency is heaven. It is an almost useless skill nowadays, and I don't do it any more on that level, but I sure wish ....

Racetrack?

[PPH]

Are we talking Gand Prix, Baja 1000 or stock car?

At least it'll make a crash a lot more fun to watch.

Re:Analogous to Human Memory?

[Anonymous Coward]

is this sort of keep-sending-signals-around-in-loops method at all like how human memory works?

Short-term memory uses some sort of an active feedback process. It has to change quickly and easily, so it has no choice. I suspect itinvolves a static latch rather than a loop, but you never know, there are lots of unexplained oscillations in the brain. Unfortunately it uses a lot of neurons so you don't get much of it, and tends to lose information quickly.

Long-term memory works by chemically altering proteins at synapses.

In the last year or so there was a paper about learning in rats. The scientists wired a few electrodes up to individual nerves in the rat's brain, then ran it through a maze and watched the nerves pulse. Afterwards they happened to leave the machine running and were astonished to see the same sequence of pulses as when the rat went through the maze, but replayed backwards.

Re:Plus one addressing

[Rorschach1]

I understand completely. My thing is 8-bit MCUs, and there's nothing like the satisfaction of coming up with some elegant, tight construct where not a byte or cycle is wasted. And there's a real economic incentive for that sort of optimization, too - I'm running my code on chips that cost $1.68 each, and doing things that my competitors might use a $6 ARM chip or $20 Java module for.