I do at times wonder just how far we can go with storage innovation, you often hear about stuff that initially has great promise, but then fails to materialize due to some gotcha.
Looking in the MIT Technology Review, there is an article from a couple of days ago, where Seagate appear to have something in the pipeline that looks very viable. They are calling it “heat-assisted magnetic recording”. Now if you are thinking “???”, well it is like this … they simply heat the magnetic regions on a disk that hold individual data bits, and so that allows those regions to be made tinier, thin enough to hold 1Tb per inch.
Here is what the article says …
“One of the most exciting things about heat-assisted magnetic recording is that it’s in its infancy,” says Ed Gage, principal technologist of heads and media R&D at Seagate. The company is targeting 2015 for its first commercial product featuring the technology.
Today’s hard disks are made of magnetic cobalt-platinum alloys. Each bit is stored on a tiny area with a magnetic field pointing in one of two opposite directions, denoting a binary digit 1 or 0. The smaller these magnetized areas are, the higher the density of the disk. When the areas get down to 25 nanometers to a square side (corresponding to 1 terabit per square inch), they become unstable, meaning that a small amount of heat can make them flip their magnetic field direction.
More-stable magnetic materials, such as iron-platinum alloys, are available, says Mark Kryder, an electrical and computer engineering professor at Carnegie Mellon University and previous CTO of Seagate. However, to write on them requires magnetic fields much larger than those conventional recording heads can produce. If, however, you heat the material, smaller magnetic fields will work. So heat-assisted recording involves heating iron-platinum disks with a short laser pulse when the head applies a magnetic field to write data.
What makes this one stand out for me is that it is not … “Here is an interesting lab-bench discovery that might one day …” but rather “Here is exactly how this works, and our first commercial offering will be in 2015“. So the chances that this will indeed pan out are quite high.
It’s not a done deal, they still have a lot of work to do, and yet it looks both viable and achievable.
As for what you might do with a 60Tb disk, now that is another quite separate discussion.