'Racetrack' Memory Could Replace Hard Drives? 149
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."
This sounds.... (Score:3, Informative)
Re:This sounds.... (Score:2, Informative)
Re:Sounds like... (Score:4, Informative)
http://en.wikipedia.org/wiki/Bubble_memory [wikipedia.org]
Re:I've seen it in fibre before... (Score:5, Informative)
The basic technique is even older than that. Google "Mercury Delay Line" for early examples: they'd make a long thin tube of mercury with transponders at each ender. It was around 5 ft per K, IIRC.
there != their (Score:4, Informative)
I will stop now before I make a simple grammatical error myself.
(yes, I know you're looking, hmm, hmm, must be one here somewhere)
Re:Anything (Score:4, Informative)
I know you meant that as a joke, but...
You should take a HDD apart some time. Though manufactured to incredibly small tolerances, they only really have two moving parts - the platters, and the head assembly (which despite having a lot of sub-parts, moves as a single unit).
And aside from them, you don't even have that much else that goes into a HDD - usually two air filters (one for keeping internal air clean, and one that balances external air pressure changes); the body itself (just a big aluminum block with an airtight lid); A magnet assembly for moving the heads; and the electronics on the visible external board. Sometimes you have one more small mechanical bit that doesn't seem to do anything (perhaps it parks the heads for shipping?); And that about covers it.
Re:Anything (Score:2, Informative)
The more things change, (Score:3, Informative)
The more they stay the same.
For those who don't know, delay line memories [wikipedia.org] have been around for at least 50 years...
Kind of interesting that they are using an old concept with new technologies.
Plus one addressing (Score:5, Informative)
The optimization was great fun, my favorite part. You could make programs scream if you paid attention.
Re:I've seen it in fibre before... (Score:3, Informative)
10Gbps = 1.25GB/s
c = 300,000Km/s (2sf)
Does 100km in 1/3s
1.25GB/s * 1/3 s = 0.416GB
I think your answer is off by a factor of 1000 (or maybe 1024)
Re:Anything (Score:4, Informative)
In other words, the guys who've been designing hard drives for the past few decades aren't stupid.
Re:Sounds like... (Score:3, Informative)
Re:Anything (Score:4, Informative)
Very low-end flash memory has that kind of write cycles. And it's typically limited to NOR flash, which is used only for code memory and limited data store due it its large cell size (largest NOR flash chips are around 256MB). Even so, Intel's StrataFlash had write lifetimes of at least 100,000 erase-write cycles, and most flash chips are underrated by an order of magnitude.
Modern bulk-dsta storage flash is NAND flash, which due to its smaller cell size (partly due to its design, and partly due to operation), means 16GB (byte, not bits) per chip is starting to become practical. NAND flash is faster erasing and writing than NOR flash, but much slower (order of magnitude) slower at reading. Plus it's I/O based - you can't "boot" from NAND flash like you can from NOR. (Write/Erase/Reads are on the order of microseconds for NAND - typically 100-500uS for write/erase, and 10uS for reads. For NOR, writes are typically 300-1000milliseconds, erases 1000ms, but reads on the order of 100ns or less).
Because of the operational characteristics of NAND flash, it typically has a 100,000 write-erase cycle limit at the minimum, with most offering at least 1,000,000 cycles (and typically lasts an order of magnitude more).
Wear-levelling algorithms and bad-block handling increase the time between writes and erases to the point where it almost isn't a consideration anymore - when the drive dies eventually, it'll really be timeto change it. And at least when an SSD dies, it dies on erases and writes, and very rarely on read. So if you get write errors, you still have a great probability of recovering all the data (except the data which was just written).
It's write-erase cycles, because erasing turns "0" bits into "1" bits. Writing turns "1" bits into "0" bits. Within certain restrictions, you can do multiple writes to a block (turning "1" bits into "0" bits, but you can't turn a "0" bit into a "1" bit without erasing), but those don't count towards write-erase cycles. (This behavior is often exploited when marking blocks as dirty and such). And they only fail on writes or erases due to internal timeouts (each cell takes progressively longer and longer to erase and write). Reads can be considered as never failing.
Re:Hello core memory? (Score:1, Informative)