It was good-bye and good riddance for IBM's hard disk drive (HDD) manufacturing business last week as Hitachi Global Storage Technologies (HGST), the company formed when Hitachi agreed to purchase IBM's HDD operation last spring, officially opened for business. Before the purchase, IBM's HDD division had lost nearly $500 million in 2001, and although the final numbers for 2002 aren't yet in, they're not going to be good. In the third quarter, IBM reported a loss of $142 million in the HDD division.
Most of the speculation about the impact of IBM's withdrawal from the HDD market has focused on the effect the deal will have on the corporate computing space. The purchase reflects ongoing consolidation in the HDD market, which currently leaves only a couple possible combinations of large companies. With IBM now concentrating on Storage Resource Management (SRM) solutions, some industry observers believe that the move toward standards-based storage such as the Common Information Model (CIM) will accelerate.
But Hitachi threw a significant challenge at the industry at the launch of HGST: The company announced its new Microdrive product, with a 1" form factor and a 4GB storage capacity. The new device builds on IBM technology first unveiled in 1999 but has four times the storage capacity of IBM's current Microdrive.
The Hitachi Microdrive, which will be available in the fall of 2003, has overcome many engineering and technical hurdles. The read/write head is half the size of its predecessor, decreasing the height at which the head flies over the disk platter by 40 percent. The number of tracks per inch has increased dramatically, and the drive has an areal density of 60 billion bits of data per square inch. The areal density was made possible by a media technology that Hitachi calls Pixie Dust. The media consists of three-atom-thick layers separated by the element ruthenium (similar to platinum). Technically known as antiferromagnetically coupled media, the ruthenium magnetic layers provide ultrahigh recording densities while maintaining data integrity.
The Microdrive announcement has several significant, long-term implications. First, high-density, small-form-factor HDDs are emerging as the enabling technology for a lot of cool new mobile and handheld devices. Apple Computer's iPod MP3 player is built around Toshiba's 1.8" disk drive featuring 10GB to 20GB of storage. Microsoft's Xbox video game player has a hard disk, and Sony plans to add a hard disk to its game player. Home media servers such as TiVo and ReplayTV might also take advantage of small-form-factor disks. Another area in which the new technology will have an impact is in the car stereo market. Consumers want to be able to store 18 to 20 hours of music on a disk, a desire that isn't lost on car stereo manufacturers. And the next generation of PDAs, digital cameras, and cell phones are going to be able to store a lot more data than they currently do.
An increase in the amount of data that personal devices store will significantly affect storage administrators within the enterprise. Their challenge will remain convincing personal device users who store a lot of data on their PDAs to back up their information safely in central repositories. Data synchronization could become a nightmare. Knowing what data is stored where, when, and by whom will become an increasingly complex task. Storage administrators will need to carefully define their storage policies and devise strategies to ensure that users adhere to the policies.
As 1" and 1.8" hard disks make their way into consumer electronics and other personal devices, vendors will realize the manufacturing efficiencies that mass production makes possible. As a result, costs will continue to drop. In the near future, small hard disks will be virtually everywhere data is found.
As with the rest of the IT infrastructure, storage technology tends to alternate between phases in which the industry market leaders drive toward centralization and phases in which decentralization becomes of primary importance. The IBM-Hitachi deal has generated comment that mostly centers on the implications for centralized storage. But the acquisition will also advance how data is stored on devices just beyond the network. Managing that data will prove challenging in the years to come.