Storage Scheme Pushes Gigabyte Capacity

Welcome Guest. | Log In| Register | Membership Benefits

August 24, 1999 (9:03 AM EDT)

Building on research done at Keele University, start-up Keele High Density aims to commercialize a memory technology it said could store 2,300 gigabytes of rewritable random-access data on a device the size of a credit card. That is equivalent to an areal density of 86 Gbytes per square centimeter.

The memory system exploits the storage properties of a new family of metalalloys, said Mike Downey, managing director of Cavendish Management Resources, in London, a venture capital firm that formed Keele High Density as a joint venture with Keele University. The company said it plans to license the technology to computer and storage companies. Downey described the memory system as solid-state and three-dimensional, with data stored in multiple layers.

The company said production costs for a 2,300-Gbyte credit-card-sized memory will be less than $50. Keele also talked of wristwatches with a memory capacity of more than 100 Gbytes.

The technology arises from work led by Ted Williams, now a professor emeritus at Keele. Williams has applied for patents on plans for optical, magneto-optical, and multibit-per-storage-site memories that promise to raise data storage densities by a factor of 20 or more.

Ironically, Keele University closed down its electronic engineering department, where Williams worked, but now stands to earn licensing revenue from the technology developed there.

Downey said many details of the technology were being kept secret as Keele High Density enters into sensitive licensing talks with a number of computer and data-storage companies. Downey declined to name any of the potential licensees.

He said the metal alloy Keele uses is akin to one being researched at IBM.

"I believe IBM has demonstrated the use of up to 10 layers using a similar material, but we don't use that many [layers]," Downey said. "That's not the key. What Professor Williams has done is found a new way to store, retrieve, and erase data."

Because the technology uses hardware from existing memory systems, implementation could be done quickly and at low cost, he added.

The quoted areal density and available optical resolution imply multibit-per-site data storage.

"In essence, that's what we have done," Downey said.

Williams confirmed the multibit-per-site nature of the storage technology, but said the principle was being applied to both conventional magneto-optical materials originally developed by Philips and Sony, and to a new alloy material of his own invention. Williams declined to elaborate on whether the new material displayed a magneto-optical or a different storage effect.

"We have built a crude demonstrator based on a credit-card format," Williams said. "We plan to build a large-scale prototype in the middle of next year with one or more licensees. That will be based on conventional magneto-optical material, although the principle of the invention is applicable to other modes of storage."

Although the system is described as solid-state, it does require moving parts to roughly align an optical system above the storage medium and to focus the beam used for writing, reading, and erasing data. Focusing is used to provide fine control of the addressing in "x", "y", and "z" directions, with the variable focus and transparent microscopic layers used to store data in three dimensions. The data-access time for the new storage technology is predicted to be around 100 Mbps.

According to the company, an additional advantage over existing data-storage systems is that only 20 percent of the total capacity is needed for error correction, significantly less than the 40 percent now needed for hard disks and the 30 percent needed for optical storage.

If the technology were to take hold, Keele said it could give rise to portable computers capable of storing as much data as 350 present-day PCs. Theoretically, such systems could be on sale within two years, the company said.

Williams has a strong pedigree in technical innovation. Between 1978 and 1982 he led the team that developed the nuclear magnetic resonance imaging, or MRI, medical body scanner.