-->

KMWorld 2024 Is Nov. 18-21 in Washington, DC. Register now for Super Early Bird Savings!

  • August 15, 1997
  • News

Letter To The Editor: "Far From Dead"

My congratulations to Tom Dale of Cambridge Consulting Group (McLean, VA, http://www.c-imaging.com/ccg) for "attempting" to make a case for the demise of optical storage in “Point,” June 16. It is difficult to take a position and ignore logic and marketplace facts. But he did a good job of it.

More than 20 years ago we announced the death of microfilm and microfiche. Sales continue to grow in a predictable, steady manner. In fact, one of the most intriguing products at the '97 AIIM show was Sony's (San Jose, CA, http://www.sony.com) announcement of a computer-assisted microfilm system. Because of the widespread installed base of microfilm and microfilm systems the unit has excellent potential.

Fifteen years ago when optical started to make its initial onslaught of the marketplace we forecast the demise of tape. Ditto 10 years ago when CD-ROM entered the arena. Surprise! Sales have expanded, technology/capacities have steadily improved and the tape market remains very strong for a majority of the players.

When CD-ROM technology was introduced industry pundits also forecast the death of the floppy. But according to Imation (Oakdale, MN, http://www.imation.com) officials more than 90 million 1.4MB floppy drives will be sold this year and the company will produce more than four billion floppies.

With the imminent roll-out of the DVD drive/media people are forecasting the death of CD-ROM technology. This year more than 66 million CD-ROM drives will be sold adding to the present installed base of 160 million. More than two million CD recorders will be sold as will more than 200 million pieces of CD-R media. In fact, according to several respected industry analysts and manufacturers the DVD installed base will be less than 50 million units in 1999 and more than 210 million CD units. Even in the year 2000 the installed base of CD will be more than double that of DVD.

Why? CD-R and CD-E technologies are getting easier to use, prices continue to drop and to be honest, there is a tremendous market for sub-1GB storage.

But what about MO? Capacities are presently 2.6GB--comparable to Mr. Dale's computer's 2GB drive. Late this year 5.2GB drives and media will be available and keeping pace with Moore's computer law, in less than 18 months capacities will be 10.4GB. These advances will be available at the same, if not lower, cost to the buyer. Storage street price of today's direct-overwrite MO is under four cents/mb and this will come down.

LIMDOW (direct-overwrite) technology is now being implemented by every major optical drive manufacturer. This has opened up new markets for the technology rather than limiting it to niches as Mr. Dale notes. Increasingly, organizations are using it for data streaming, multimedia and Web page storage and audio/video pre-production and production.

While Mr. Dale is correct that RAID is faster for retrieval, VARs and users also realize that while the name stands for Random Array of Inexpensive Drives, RAID is not inexpensive. If it is true, as Santa Barbara-based Strategic Research (http://www.sresearch.com) points out, that 80% of the data stored on a network is inactive, why would any budget-conscious network or IS manager store that data on his or her RAID system? Forrester Research (Cambridge, MA, http://www.forrester.com) and Gartner Group (Stamford, CT, http://www.gartner.com) both point out that less than 20% of a corporation's information and knowledge is stored on-line so that it is available to decision makers.

As an MO company official noted at the recent Strategic Research Network Storage Conference, they are enjoying excellent sales success by moving all of these paper documents to optical disks and optical jukeboxes so they are available to network and enterprise personnel. IBM's (White Plains, NY, http://www.ibm.com), EDS's (Plano, TX, http://www.eds.com), Oracle's (Redwood Shores, CA, http://www.oracle.com) and other integration firms are having excellent success in helping firms develop, install and manage data marts and data warehouses that make document, data and image knowledge available to decision makers who must now deal with global product development, manufacturing and sales issues.

Few company managers can cost justify putting 100GB to 1TB+ worth of mission-critical but seldom used data on RAID. It is much more cost-effective, much safer and almost as fast to retrieve when this information is stored on optical media in optical jukeboxes. Software developers have done an excellent job of adding features and capabilities to sort/retrieve and jukebox management software so that in less than five seconds an individual anywhere in the world can have all of the data he or she needs.

Just as Mr. Dale purports that optical storage is deceased, Terastor (San Jose, CA, http://www.terastor.com) and Quinta (San Jose, CA, http://www.quinta.com) are developing some intriguing optical technology that could produce products that store 20GB to 40GB of data on a single disk. Both firms are very well funded and have outstanding engineering teams. In three to five years they could introduce a totally new generation of MO storage.

Perhaps that's too long for Mr. Dale but it closely parallels the path of DVD, which we have to remember was developed with the idea of replacing your VCR system, not for storing data.

Direct-overwrite LIMDOW technology breathed new performance life into MO storage. Members of the Optical Storage Technology Association (OSTA) (http://www2.osta.org/osta/default.shtml) have committed to moving forward more aggressively on the introduction of higher capacity products and in providing seamless data interchange between drives.

In the last 30 years we have produced more information than we did in the previous 5,000 years. According to futurists we will double the volume of information we receive every five years. Are you going to store that information on fragile hard drives? Are you going to copy it to tape where data life has increased from two years to ten? Are you going to store it on optical which has a minimum data life of 30-50 years and a very excellent price/performance ratio?

The answer will probably be a mixture of these technologies and they will all be "hooked" together/managed by some very excellent storage management software that is now available.

Rather than "Will RAID eat your WORM?," Mr. Dale should check his glass again...he may have swallowed the worm.

Andy Marken
LIMDOW Consortium

KMWorld Covers
Free
for qualified subscribers
Subscribe Now Current Issue Past Issues