X

Network storage's hazy future

Analyst Ashok Kumar says the business is about to get rocked by a technology shift, but too few companies are paying attention to the writing on the wall.

3 min read
Customers who understand network storage no doubt want to exploit its benefits. But there are still many more questions than answers surrounding both the technology and the future of the business.

Consider, for example, Brocade Communications, the clear industry leader in the switch-niche segment of the network storage industry. Brocade has historically been highly valued. It is the early leader in a market with enormous potential. And it enjoys a solid position, with its products included in most of the storage area network packages for Fibre Channel switches.

However, that is not the same as saying Brocade is on the fast track to generating billions of dollars anytime soon. The company always seems to be on the verge of making bigger numbers with some new, untested product in a market suddenly defined as the center of the action.

That makes for interesting reading, but investors should realize that undefined growth markets often fail to materialize.

To be fair to Brocade, the company's Silkworm product line appears to generate a reasonable business. Still, one gets the distinct impression that the revenue performance of Silkworm has been a slight disappointment. Although nobody would ever state that in public, there's a reason why some things get emphasized more than others.

The product du jour now is the 12000 switch. It is positioned at the high end of the market as a director-class product, even though it is more of a collection of switchblades than a designed-from-the-bottom director. Compared with the competition, the 12000 offers few advantages, other than the fact that it's a Brocade product and will work with other existing Brocade switches. But that raises an interesting question: Why should a bunch of directors have to be tied together at all?

The 12000 is a little bit less than a me-too product that showed up late for the dance. This is not anything to get excited about unless you happen to work for Brocade. One of the reasons for buying a director is to avoid having to combine switches and suffer the additional administrative headaches that go along with increased complexity. In a sense, a move to directors is a move away from legacy Fibre Channel switch interoperability. The point is simple: The 12000 is a little bit less than a me-too product that showed up late for the dance. This is not anything to get excited about, unless you happen to work for Brocade.

There's something more interesting going on here. Considering how customers are clamoring for the products, the fascination with director-class products by Fibre Channel switch companies is not difficult to understand. The early adopters of Fibre Channel include many of the financial services companies that provide analysis and investment coverage on the industry.

Unfortunately for the Fibre Channel switch industry, this fascination with high-end products suggests an industry just asking to get rocked by a disruptive technology. The early crop of sellers often overshoots the needs of a broader mass market. Instead, they create "over-functioned" products designed for early-adopter customers with unusually extreme requirements.

Fibre Channel will survive as the input-output solution of choice for the tip of the iceberg, but Ethernet storage is going to get the rest of the business. Director-class products are likely to overshoot the sweet spot of the network storage market. The real disruptive technology in storage networking is input-output based on Ethernet and TCP/IP. While the Fibre Channel switch industry has screwed itself into a mouse hole for high-end, expensive products, the Ethernet industry has made continual progress with a disruptive technology that will undercut the cost of Fibre Channel operations by a nontrivial margin.

Yes, Brocade is the current storage area network switch leader, but the company does not appear to acknowledge the seriousness of the threat posed by Ethernet storage technology. It is not simply a matter of comparing the costs per port, which Ethernet wins hands down, but the availability of expertise for Ethernet and TCP/IP that makes the big difference.

Fibre Channel will survive as the input-output solution of choice for the tip of the iceberg, but Ethernet storage is going to get the rest of the business. If Brocade can figure out how to take its leadership into the Ethernet storage industry, it might be worth its current high stock multiple.

But the Ethernet industry remains oblivious to the upstart's attempt to kick its hindquarters. In the meantime, Brocade appears to be pouring on the afterburners as it accelerates into the vanishing point of an overpriced technology.