Did we learn anything last year?
By Marc Farley
A year ago, the storage industry was trying valiantly to hang on to the myth that it was recession-proof. It did not take long to figure out that sales volumes were off and that things were going to be different than we expected. But it took most of the rest of the year to find out how low things could really go. The events of September 11 certainly had a severe impact, but it is pure denial to ignore the fact that the industry tanked in 2001.
If there was anything to learn from 2001, it was caution in misreading the tea-leaves regarding storage business trends. The recent Enron debacle serves as a potent reminder that individuals in companies often put the best spin possible on their situation, even when they have "no visibility" to the market trends.
Take your pick of storage companies and consider whether or not their current messages are simply echoes of last year's optimistic spins.
2001: The year of empowerment for IT administrators?
Another important lesson was that IT administrators actually do have some control over the growth of stored data. The recession-proof concept was based on the notion that administrators have no control over data storage and that users continue to force the issue with undisciplined fervor. Instead, what we found were IT professionals becoming vocal about their need to unlock the trapped capacity in under-utilized subsystems. This was a far cry
The truly encouraging news in this is that consumers of storage are taking a more active role in understanding their storage needs and architectures and in letting companies know their requirements. This is an extremely important step in the evolution of the storage industry. As more customers make direct contacts with more storage companies and vendors, the potential for broadening the channel increases, which translates into greater opportunities for more companies and less costs for purchasers.
But for now, don't hold your breath, the storage industry is still dominated by OEM sales opportunities and will continue to be for several more years.
Storage in 2002: Storage still follows systems, DAS a strong factor
So how should we think about the industry at the start of 2002? If we take our cues from recent history, we should continue to be cautious. The dangerous myth now is that there is pent-up demand for storage that will result in a flood of orders once the economic skies brighten.
This is a myth because it ignores the fundamental dynamics of storage purchases. History tells us that storage sales follow system sales. The storage industry doesn't like to hear that message, but I'll argue that history tends to be more accurate than prognostications. The operating reality in IT is that nobody wants to upset the flow of operations if they don't have to. This means other techniques will be found to keep things working before significant storage restructuring occurs.
In general, people are reluctant to introduce new technologies into their workhorse systems environments. The risks of having self-imposed downtime, or worse, loss of data, are not worth the effort of fixing something that isn't really broken.
This does not mean that new storage technologies aren't extremely important, because they are, but it's simply a statement that progress happens when the opportunities are right. Storage consumers are risk averse. The perfect time to change storage architectures is when systems are being changed, or when new systems are added and there is minimal risk in making the change.
New technologies can be built-in to these systems, without having to take them down or risk incompatibilities. The analogy isn't the best, but people who ride trains don't change trains by jumping from one moving train to another -- they wait until the trains have stopped at the station and then board.
The shift from today's prevalent DAS architectures to tomorrow's SANs or NAS is going to take many years. Until consumers of storage are more comfortable with the technology, it will take longer than anticipated to overhaul storage infrastructures. This is certainly true for SANs, but it also applies to other new storage technologies too, such as DAFS. As much as I like the concept of DAFS and what its potential is, it requires an infrastructure change at the customer site, and therefore it will be a slow and steady ramp to success, not a hockey stick.
I believe the long-term picture for the storage industry is better than ever. The dynamics are shifting to storage independence from systems. But, in the short term, systems are still driving storage results. The industry will not see its next period of fast growth until the server manufacturers see theirs.
About the author: Marc Farley, is the author of Building Storage Networks, 2nd edition, often referred to as the "industry bible" for understanding the basic concepts behind storage networking. Farley is also a resident expert on searchStorage, in the Ask the Experts category of Storage Networking: A to Z.
This was first published in January 2002