News & Analysis

Comment


Ryan Tennill

10/13/2011 6:00 PM EDT

I don't know that this is really true. Look at the pricing of SSD which provide ...

More...



sharps_eng

10/11/2011 5:49 PM EDT

Interesting point about 2 dollars per disk being a limiting factor for ...

More...

Drive makers forge HAMR for 2015

Rick Merritt

10/11/2011 9:00 AM EDT

SAN JOSE, Calif. – Hard disk drive makers are gearing up for a generational shift to heat-assisted magnetic recording (HAMR) technology. Leading researchers will share their progress toward that goal at the annual Diskcon event next week.

For years, drive makers argued over the future of their road map. Some, led by Seagate, lobbied for HAMR; others, led by Hitachi GST, called for a move to bit-patterned media.

Both technologies aimed to deliver drives that could pack multiple terabits of data on a square inch of disk space. Today's perpendicular recording techniques are expected to run out of gas at an aerial density of one to 1.5 terbaits per square inch.

Toshiba will claim at Diskcon it has found a way to use perpendicular recording to pack a terabit of information on a square inch of disk. It will demo 2.5-inch drives with 500 Gbyte platters at that density.

Early this year the two sides in the road-map debate quietly converged on HAMR as their next step.

"There's a general consensus the huge shift beyond perpendicular is at least three years out, so mainstream products won't ship until 2014 or 2015," said Mark Geenen, president of IDEMA, the disk drive trade group and host of Diskcon.

"Sometime before that—as early as 2013--companies may do a soft launch of the new technology to test it out in a less grueling areal density environment--it’s typical one or two companies will do it earlier than necessary," Geenen said.

Over the next year or two, drive makers are expected to use shingle magnetic recording, a variant of perpendicular, to push aerial density to or slightly beyond a terabit.

"But the big shift appears to be a consensus on heat assisted being first and in the future moving to bit patterning--it’s a pretty big set of challenges," he said.

HAMR's challenges include finding the right recording materials, then solving a range of engineering challenges such as how to integrate laser diodes and recording heads. The technique uses the diodes to briefly heat a tiny area on the disk so its data can be read.

While difficult to engineer, it is now seen as a clearly easier step than the main alternative--patterning multiple terabits of data uniformly on a platter. Proponents have yet to demonstrate ways to cover a full disk with tiny magnetic dots in a way that can be mass produced and adds no more than two dollars to the cost of a disk.

Patterned media is believed to require billions of dollars in new capital equipment. "It isn't even on the radar" for drive makers today, Geenen said.

However patterned media is expected to make a comeback when HAMR runs out of gas. But that may not be until 2020 or beyond when drive makers are at an aerial density measured in multiple terabits, Geenen said.

Last year drive makers formed the Advanced Storage Technology Consortium (ASTC) to pool resources needed to make the generational leap beyond perpendicular recording. The 13-member group is lead by Hitachi GST, Marvell, Seagate and Western Digital.

In July the group started funding pre-competitive research at universities and institutes, mainly on HAMR though some money is going to patterning work, too. The group has not disclosed its budget, but it is measured in "several million dollars," Geenen said.

ASTC is expected to expand eventually into other efforts such as industrial research.





sharps_eng

10/11/2011 5:49 PM EDT

Interesting point about 2 dollars per disk being a limiting factor for high-density storage. If that could be put outside the raw build cost of the drives, and simply added on at the end when you bought your $100 drive, no-one would object to paying that 2-dollar licence fee or premium to get double the density or more. But, when you stick two bucks on the raw build cost, and shove it up through 5 or more layers of sales distribution, it becomes 100 bucks extra, and no sale. So it doesn't matter how big a jump you could get in performance, it has to be cheap enough to hold the price the same as before.

Sign in to Reply



Ryan Tennill

10/13/2011 6:00 PM EDT

I don't know that this is really true. Look at the pricing of SSD which provide much lower storage/$ but are still making volume sales due to performance. If disk IO performance can be said to scale with density for rotational drives then there is a good opportunity to use performance/$ pricing rather than density/$ while manufacturing and R&D costs are amortized.

Sign in to Reply



Please sign in to post comment

Navigate to related information

Featured Job On
Scroll for More Jobs

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)