Fujitsu pushes limits of data-storage technology

Researchers in Japan have reported success in an advanced data-storage technology that could help yield hard-disk drives with capacities of seven times or more than today's most advanced drives in as soon as five years.

Free IT resource Virtualization Insights from Top Experts - Learn how virtualization gets real! Sponsored by Dell Free IT resource Try Sun servers, workstations and storage products free for 60-days. Sponsored by Sun Microsystems

Their work is a refinement of perpendicular storage technology, a method of data storage that is only just beginning to come into commercial use in hard-disk drives. Drive makers are switching to perpendicular storage because it allows much more data to be stored on a disk. This is because the magnetic particles on which data is stored stand perpendicular to the disk's surface and so more of them can be packed onto the disk than in the current longitudinal recording method in which they lay flat.

The new research further increases the storage ability by organizing finer particles of magnetic material into a fixed, regular pattern, said Kenichi Itoh, director of science and senior research fellow at Fujitsu Laboratories' storage intelligent system laboratory. Itoh is working on the project with colleagues from Yamagata Fujitsu  and the Kanagawa Academy of Science and Technology and the work is supported by the Japan Science and Technology Agency.

Researchers start with a piece of glass and a layer of aluminum is added to one side. This sheet is put through a process called anodization, which is one of the keys to the new technology, in which electricity flows through sulphuric acid from a negatively-charged cathode to the glass-aluminum disk, which acts as the positively-charged anode. The process, which takes about 90 seconds, results in numerous minute holes called nanoholes being formed in the aluminum. Each nanohole is about a thousandth the width of a human hair.

Typically the nanoholes appear at random in the aluminium. However, Itoh's team has been able to get them to form in a uniform pattern by stamping the aluminum with a die before annodization. The result when viewed under an electron microscope looks similar to a honeycomb pattern.

Next the holes are filled to just over the top with cobalt, a magnetic material, and this is polished to give a smooth surface. Before the disc is finished, a protective layer is also added. The result is a disk covered with billions of tiny cobalt-filled holes each of which can hold a magnetic charge, forming the basis of a high-density data storage disk.

"This discovery may open the way for 1Tb per square inch in density in perpendicular recording media," Itoh said. Today's most advanced drives can store somewhere between 120Gb and 140Gb per square inch.

The amount of data that can be stored in a square inch of disk space is a critical measure for hard-disk drives. The disks are a standard size -- typically from 1-inch in diameter through 1.8-inches and 2.5-inches to 3.5-inches. Increasing the capacity of drives by enlarging the disks is out of the question. So drive makers are usually faced with two options: either stack two or more disk platters inside a single drive, or squeeze more data onto the disk. Adding platters is technically easier, but increases the size and weight of the drive. The number of platters that can be added is also limited because the thickness of the drive, like the diameters of the platters, has to fit a certain standard.

Itoh's technology won't appear in commercial drives anytime soon. In the lab, his team has managed to prove the technology by forming patterned nanoholes in an area 3 millimeters square, although typically the work is done on much smaller areas. On 2.5-inch diameter disks of the type used in laptop computer drives, Itoh has managed to form nanoholes, although these have been random rather than in the regular pattern needed.

To scale up the small squares of organized nanoholes to the size of a disk requires advances in other technology, including electro-beam lithography equipment that can work at a finer resolution and over a larger area.

There are also several other technologies that need to be refined or developed before such can drives appear, he said. The servo control technology that is responsible for moving the disk head across the media needs to be improved to work at finer steps, and the drive heads themselves need to be improved and better signal processing technology developed.

However, Itoh is confident that these hurdles can be overcome, possibly in as little as five years. "I don't know exactly how long it will take but it will come," he said.