Download the authoritative guide: Enterprise Data Storage 2018: Optimizing Your Storage Infrastructure
Recently, EMC -- and I do not usually or like to mention vendors -- posted a video and a slogan used at a meeting that "Tape Sucks" So for almost 20 years that I am aware of, and many more if you talk with people older than I am, tape has been a dead technology. Why do some storage vendors keep saying that tape is dead when clearly it is alive and well and is actually growing significantly for archival data? Again, this goes back to my question: Do they think we are stupid, do they think if they say it enough, we will believe it? Maybe do they think we have not evolved enough to remember what happened last time or that information is not passed down from generation to generation?
Holographic Storage and other tape and optical replacements
In the early 1990s millions of dollars were put into holographic storage research and the technology was touted as being five years away. In the late 1990s, a company called Terastor said that it was going to replace tape, CD-ROM and lots of other storage media. I visited with Terastor two times as part of a technical evaluation we were doing for a customer. I thought that the technology was interesting and was impressed with their people.
My understanding is that they got a few of writer/reader and media pieces working, and took them to shows but could never manufacture them. From reading the press reports back when they went out of business, they burned through more $800 million dollar with very little to show except some patents. Honestly, it was very cool stuff if it had worked. The same goes for about six years later and InPhase. They were going to make a holographic disk. This, of course, was different technology than Terastor, but from the point of view of replacing tape, and other optical they had the same goals. I visited InPhase for two customers. They had a great story of how they were going to make the product and lots of smart people, but they did not ever get the product into production and I never thought they would or thought they could be competitive. Everyone that attended the meeting really wanted them to succeed and some were hopeful. I was not (a curmudgeon before my time). I said to myself been there, done that.
Tell me the point of this series again?
I have been thinking a lot about why we seemingly cannot learn from the past. Is it because we think someone will be successful where others have failed, or is it that we beat our chests because we think it will change the reality that we are facing? Technology changes require that many things be considered for that change to be successful. The tape example is a good one. Up until very recently with the release of new enterprise tape from IBM and Oracle disk drives had greater density than tape, yet when these high-density products came out it made no difference to the mantra from parts of industry that tape is dead. This sounds like politics somewhat: If you misstate the truth often enough, people will believe you on the right or the left. Long term, we need to get out of the mode of not questioning bold statements without understanding the details and information behind those statements.
I have a saying: Minutia matters. It does not matter all of the time, but matters more often than people seem willing to believe. I rarely accept things at face value and always try and analyze and understand why people take a position or make a statement. I have argued with the tape-is-dead crowd for almost 20 years. When I provide all the facts, they never seem to want to address all the facts, but only the facts that they want to pick and choose. You can pick and choose facts to come to a conclusion, but the market choose that way. Technology, with rare exceptions, does not follow straight lines for long (Moore's Law being the rare exception). We live in an ever-changing and complex world, a world in people seemingly do not want the facts to get in the way of their opinions.
Next time, I will address the same issues for the server and file system side of the equation. Remember the minutia matters (at least a good percentage of the time).