EXECUTIVE SUMMARY

This report reviews the status of Japanese high-density data storage technology, manufacturing and R&D in comparison to that in the United States. It covers various optical data storage technologies, magnetic hard disk drives, and magnetic tape drives. Information sources used in the study include literature review, visits to 15 relevant sites in Japan, a review of the draft reports by panelists, site visit hosts, and study sponsors. The report also includes information obtained from presentations from U.S. panelists and Japanese counterparts given at a workshop that took place in Tokyo prior to the site visits.

The panel found that Japan clearly leads in optical data storage and magnetic tape drives, the United States and Japan are competitive in magnetic hard disk drives, and researchers in both countries are clearly aware of future challenges in data storage technologies for the next decade. Both the United States and Japan put significant emphasis on long-range alternative data storage techniques.

There are clear possibilities for Japan to surpass the United States in the hard disk drive segment and opportunities for the United States to catch up in optical data storage. Japanese government funding in data storage in general appears to far outstrip comparable government funding in the United States. In optical data storage, with Japan being in a comfortable leading position, most R&D funding is derived from industrial sources, although new large government programs in this area are being started. However, there is substantial present government and industrial support for R&D in magnetic hard disk drives where Japan trails the United States.

MISSION AND PROCEDURE

The purpose of this WTEC panel study of high-density data storage in Japan was to investigate and report on the current status of Japanese data storage technologies compared to that in the United States. In particular, the panel's mission was to assess status and directions of Japanese R&D in high-density data storage technologies, and to contrast them with approaches taken by the United States.

To meet this goal, the panel researched the professional literature of both countries, visited numerous companies and government organizations in Japan, attended workshops on emerging technologies in the United States, and organized a workshop in Japan with participants including academic researchers. This report is a synthesis of the panel's findings reached by identifying general directions that transcend institutional boundaries.

The WTEC panel focused its investigations on four main areas:

  1. optical data storage technology
  2. magnetic hard disk technology
  3. magnetic tape technology
  4. emerging long-term data storage technology alternatives

The organizations chosen for site visits have established a leadership role in some important aspect of data storage technology. In all, the panelists visited 15 Japanese institutions including five companies and one government organization involved in optical data storage only, four companies focused on magnetic data storage only, and five companies that are heavily involved in both magnetic and optical data storage.

OPTICAL DATA STORAGE

The recent economic recession in Japan that affected most areas of heavy investment seems not to have altered the strong commitment of the Japanese industry to optical data storage technologies, especially to technologies aimed at multimedia applications. Significant investments are clearly being made at each company that the panel visited to enable the insertion of novel concepts into new products to be launched over the next five years.

In Japan, the present focus for optical storage is to exploit its advantages to gain market segments that are not well served by magnetic hard disks. Over the next five years this market segment will grow into revenues of about $20 billion per year. About two-thirds of present revenues are derived from magnetic hard drives. Optical storage offers a low-cost, reliable, and removable medium with excellent robustness and archival lifetime. A key targeted market is to replace magnetic tape for video camcorders and VCRs. Additional applications include storing digital photographs, recording movies and other video materials, and multimedia presentations for home and business.

Over the last 15 years and especially since the early nineties, Japanese companies have heavily invested in optical data storage to the point that now the Japanese industry enjoys a comfortable lead in this area. This lead is not only in the manufacturing and R&D of conventional optical disk media and disk drive systems, but also in the manufacturing of the enabling optoelectronic components such as CD lasers. The success of the CD and new DVD family of products has resulted in a 20% annual revenue growth rate in recent years, enabling the Japanese optical storage industry to contribute significantly towards the support of the optoelectronics component industry in Japan.

It is clear that the infrastructure for manufacturing and R&D in optical data storage differs strikingly between the United States and Japan. Indeed, many big and very big companies, such as Sony, Hitachi, Matsushita, Fujitsu, Toshiba and Canon, dominate the landscape in Japan. Even more importantly, some of these companies are also involved in magnetic hard disk drive storage systems, providing their company policy makers a global outlook on opportunities in data storage. At all the sites visited, the panel sensed a very strong support, awareness and commitment from the upper management to current and future development of optical data storage technology.

In addition, there exist extensive teaming arrangements among these institutions, as well as with smaller companies, providing a formidable R&D and volume manufacturing capability (e.g., the Advanced Storage Magneto Optical (ASMO) consortium). However, these teaming arrangements also lead to competition and different standards that delay the acceptance of optical storage products by the consumer. At the same time, the creation of standards by clusters of Japanese companies makes the penetration of conventional optical data storage by U.S. enterprises nearly impossible. Cross-licensing arrangements make the matter worse.

Japan seems to be extremely well positioned to lead the manufacturing of optical data storage products over the next five years. Within this time frame a clear technology roadmap exists to push the effective areal densities of present DVD-type products up to 50 Gb/in2, targeting multimedia applications. Magneto-optical (MO) disk technology, recently re-strengthened by the ASMO alliance, is claimed to be able to surpass phase change media (e.g., DVD-RAM) capabilities within the same time frame for these applications, while becoming more and more competitive in performance with magnetic hard disk drives. With the limitations of conventional longitudinal magnetic recording in sight, it is becoming increasingly difficult to dismiss MO technology as a potential threat to the U.S. magnetic hard drive industry.

In addition, significant investment is being made towards mastering technologies to support DVD-ROM needs. To this end, a well-thought-out roadmap that leverages on microfabrication techniques (e.g., e-beam lithography) is in place to push the limits of mastering techniques to areal densities exceeding 100 Gb/in2. Finally, for emerging alternative optical storage technologies, Japanese companies seem to put a significant emphasis on probe storage that may provide Tb/in2 areal densities for certain applications requiring small size and low power. Japanese universities and research organizations work mostly in partnership with industry on advanced concepts that can be incrementally integrated into present approaches.

In contrast to the heavy investment from the industry itself, the panel learned that over recent years, there has been only limited Japanese government support for optical data storage. This small support was mostly directed towards R&D organizations and universities perhaps because of the leading position Japanese industry has enjoyed in this area. However, the panel members also learned that this coming year, a large government supported R&D program is being planned, targeting the Japanese optical data storage industry and aimed at longer term R&D. Increasingly, the Optoelectronics Industry and Technology Development Association (OITDA) seems to play a coordinating role in seeking such government funding and generating technology roadmaps in Japan.

The optical storage manufacturing effort in the United States is limited to the Hewlett Packard Company, which produces CD and DVD products, and a small company, Pinnacle, which addresses niche markets in removable storage. In addition, a few entrepreneurial companies (e.g., Terastor, Quinta, Maxoptics, Calimetrics, Call/Recall Inc., Holoplex...) each carrying on R&D in distinct revolutionary approaches, aim at 100 Gb/in2 areal densities and beyond. Also, IBM continues its long term R&D in optical storage and develops a considerable amount of intellectual property. Furthermore, there are a large number of manufacturers of optical libraries that serve niche markets. These U.S. producers of optical libraries assemble Asia-made CD drives in their systems but leverage on proprietary know-how in support software and networking.

At the present time, the entrepreneurial companies are isolated and lack large-scale manufacturing capabilities necessary to launch consumer products and may therefore be ultimately dependent on Japan's formidable OE and disk media industry. However, recent acquisitions of optical storage companies by U.S. magnetic storage manufacturers and recent U.S. venture capital interest in this area seem to signal perhaps the beginning of a trend towards the reorganization of the data storage industry in the United States for gaining better overall competitiveness.

Some U.S. universities still play an important role in conventional optical storage for both Japan and the United States since a significant amount of know-how on optical disk heads and metrology still exists at these institutions. Because little U.S. government funding is available in this area, some of this effort is sponsored by the smaller U.S. companies, but most support comes from Japanese institutions.

However, in addition to IBM, significant know-how and intellectual property reside in U.S. universities and small U.S. R&D companies on several unconventional long-term optical data storage approaches that promise data densities approaching Tb/in2. These include near field and solid immersion lens approaches, volumetric (multi-layer and holographic) storage, and probe storage techniques. In addition, in recent years, under government funding, the United States has gained an advantage on certain potentially enabling technologies such as vertical cavity lasers (VCLs), array optics, and MEMS. These powerful technologies may impact or become affected by optical data storage. VCLs and optical arrays may enable high data rate optical drives by exploiting parallelism. Micro machining can find a high volume application in either the disk drive industry or for probe storage. The United States has a slight lead in research, but a high volume application and the investment it draws could quickly evaporate that lead. Japan can easily take the lead in these areas away from the United States with its process development strength.

It is the panel's overall assessment, as summarized in Table ES.1, that Japan clearly leads in all major technical areas of optical data storage with the exception of certain R&D areas and certain products serving niche markets. Japan is out ahead in optical recording, but is struggling for a business model in which many companies become profitable rather than the two or three that attempt to control the standard. There is a necessity for the required copy protection solution as part of the removable optical storage standards. Due to the large U.S. market size and powerful U.S.-based content providers, opportunities do exist for U.S. companies to generate and arbitrate such solutions.

Opportunities do also exist for the United States to re-enter the optical data storage market via new technologies. Solid immersion lens based approaches appear promising in the short term. For the longer term and significant gain in market share, volumetric parallel accessible storage systems like holographic and two-photon multi-layer recording techniques appear most promising. The key issue is an inexpensive yet reliable write once material or preferably an erasable volumetric material. With the information explosion on the net, searching for desired data becomes a critical factor. Development of suitable hardware that exploits parallel readout to facilitate content-based data search may point to a potential opportunity. In addition, investing in micro-mechanics for micro-actuators as well as for probe storage and creating a new infrastructure in the United States to support future data storage approaches certainly appears compelling at this time.

Table ES.1
Optical Data Storage Technologies: Comparison of State of the Art in Japan and the United States

Technology

Japan

Status

Trend

Conventional optical disks - Heads

+ gaining ground

- Mastering

+ gaining ground

- Media

+ gaining ground

- Channels

+ gaining ground

- Servos

+ gaining ground

- Drive integration

+ gaining ground

- Library integration

- no change in trend

ROM media

= gaining ground

Emerging non-conventional near-field optical disk technologies - Heads

- gaining ground

- Mastering

+ gaining ground

- Media

+ gaining ground

- Channels

- losing ground

- Servos

- losing ground

- Systems integration

- losing ground

- Head-disk interface

- losing ground

Probe storage - writing and reading

= no change in trend

- Media

= no change in trend

- Processing

+ no change in trend

Holographic R&D

- losing ground

2-Photon multi-layer disk R&D

- losing ground

Supporting technologies - MEMS

- gaining ground

- Passive micro-optic assemblies

+ gaining ground

- Advanced micro-optics

- losing ground

- Mastering lasers

= no change in trend

- Optical media testers

= losing ground

Key

+

Ahead of United States

up arrow: gaining ground

Gaining ground

=

Two countries even

down arrow: losing ground

Losing ground

-

Behind United States

right arrow: no change in trend

No change in trend

However, we should point out that except for low cost, personal, removable storage, which will be a sizable market over the next five years ($20 billion/year in revenues), optical storage will not be the dominant technology for storage. Magnetic disk drives will be. The United States should not lose this technology and its future investment base.

MAGNETIC STORAGE

In spite of the recent economic difficulties in Japan, it was found that Japanese companies are gaining strength both technologically and marketwise in the area of magnetic data storage. For the past three years U.S. company market shares have decreased as a percentage of total, while Japanese company market shares have increased. Furthermore, although U.S. companies still account for over 75% of magnetic hard disk drive sales, a much larger percentage of the components used in those drives are made in Japan.

It was found that the Japanese government, companies and universities were making major commitments to research on advanced magnetic and other data storage. Twenty-one Japanese companies in the magnetic data storage area are funding the Storage Research Consortium (SRC), which currently supports 34 Japanese universities working toward demonstrating 20 Gb/in2 on a magnetic hard disk. Major funding of university research in this area by Japanese companies is a relatively new phenomenon and was originally announced as a direct response to the success of the United States in creating centers for research on magnetic data storage at universities such as Carnegie Mellon University, the University of California at San Diego and Stanford University. In addition, the Japanese government is providing 5 billion in funding to six industrial firms to develop 40 Gb/in2 magnetic recording technology by funding the Association of Super-Advanced Electronics Technologies (ASET) program through MITI. The SRC and ASET programs are coupled through the leadership of the six industrial firms and researchers, which are part of both programs.

Japanese companies were clearly aware of the long-term research issues such as barriers to increased storage density caused by thermal instabilities and had relatively large research programs addressing them. By comparison, with the exception of IBM, most U.S. companies are working only upon the next generation or two (1-3 years) of product and are relying upon their involvement in the National Storage Industry Consortium or U.S. universities to deal with longer term (5-10 year) issues.

It was the judgment of the panel that, in the area of magnetic disk drives, Japan was either gaining ground on the United States or maintaining parity in all areas of the technology. This is shown in Table ES.2, which shows the panel's judgment of the status of Japanese technology compared to that of U.S. companies. It was judged that Japanese technology in the critical areas of heads, head-disk interface and track following are already equal to those in the United States, and that the Japanese are gaining on their U.S. counterparts in the area of disk media. Without the advanced research done by IBM in the United States, the Japanese would clearly be judged to be ahead in several of these areas.

Table ES.2
Magnetic Disk Drives: Japan Compared to the United States

Component

Status

Trend

Magnetic Heads

Parity

Pulling Ahead

Magnetic Media

Trailing

Gaining

Channel Electronics

Trailing

Maintaining Position

Head-Disk Interface

Parity

Maintaining Position

Track Following Servos

Parity

Pulling Ahead

System Integration

Trailing

Gaining

Another strength of the Japanese is that they have major programs in technologies such as perpendicular recording and probe recording. Advocates of perpendicular recording argue that it may be able to achieve 5-10 times higher density than conventional longitudinal recording, before thermal instabilities, which are expected to limit the density of conventional longitudinal magnetic disk drives, become a problem. Probe recording on various media types has been shown to have potential of storing information at densities several orders of magnitude higher than possible with longitudinal magnetic recording. Canon showed the WTEC team probe recording of 10 nm sized spots on Langmuir-Blodgett type media. Most U.S. disk drive companies have no activities in either perpendicular or probe recording.

In the area of magnetic tape drives, the Japanese have a critical advantage in that they are the only source of high-performance recording tape. This leaves U.S. magnetic tape drive manufacturers critically dependent upon Japanese companies for tape media and is a serious threat to the development of future high-performance tape drive products. To develop advanced tape drives, it is necessary that all the components be developed together and integrated. Currently, as Table ES.2 shows, U.S. companies are considered to have the lead in magnetic heads, servos, signal processing and systems integration of tape drives for data storage, but if U.S. tape drive manufacturers are unable to obtain advanced tape media samples in the future, this lead could disappear very quickly, because U.S. drive development will have to wait until suitable tape becomes commercially available, which could be long after competitors have developed their products. This situation is particularly perplexing, since advanced media development, including media substrates and head-media interface, are seen as the primary roadblocks to future progress in the technology. This problem is not easily solved as there is currently only one viable tape media supplier in the United States, and it is not licensed for the double-coat technology, which was developed by Fuji in Japan and which is currently the favored approach to making high-performance tape media.

FUTURE MARKETS

Advances in data storage technology are critical to computer technology; however, advances in data storage technology have been so rapid over the past several decades that it has been taken for granted by the computer industry. Areal density on magnetic hard disk drives has advanced over 2 million times since the first disk drive, the RAMAC, was introduced by IBM in 1957, and this has dramatically reduced cost of data storage. In the last ten years alone, the average price per megabyte of information stored on disk has dropped from $12 to less than $0.1 as shown in Fig. ES.1, which plots past and projected future cost per megabyte of disk storage and total worldwide disk drive sales revenue from 1988 to 2000. This has fueled good growth in the data storage industry, but, with the advent of computer networks, the demand for data storage is increasing and the industry is projected to grow at an even more rapid pace in the future. The fact that an increasing portion of the components of disk drives are being made in Japan, the recently instituted cooperative university-industry-government cooperative research programs in Japan, and the strength of the Japanese companies in long-term research on magnetic data storage compared to those in the United States suggest that Japanese companies will assume a larger portion of this future market. Indeed, Japan has gained a larger percentage of the market for disk drives each of the last three years. The United States needs to make a larger commitment to long-term research in this area and should consider working cooperatively with Japan in areas where both sides bring something to the program.


Fig. ES.1. Trends in disk drive sales and cost per megabyte.


Published: June 1999; WTEC Hyper-Librarian