Enterprise Storage: Storage’s Endless Summer
While the growth of networked storage has seen some vendors sharing swells while others fight over the surf, in general, you as the technology consumer have benefited.
If you’ve never experienced the thrill of a "rail grab cutback" or a "360 air tail slide," then maybe your quest for the Endless Summer has been limited to nostalgic surfing movies.
Endless Summer is a code for surfers that means a life-long, year-round pursuit of the perfect "break." By extension, it also means the abandonment of all conventional cares (including working for a living) in order to dedicate oneself to an ideal of perfect balance and oneness with nature. Endless Summer is also a metaphor for a kind of thinking that’s unbounded by conventional wisdom. Surfers believe that once you leave it for the pragmatism and logic of adulthood, it’s gone forever. Or is it?
Once in a great while, you might see a coworker succeed in an Endless Summer quest. He or she sets aside the concerns of the workaday world and focuses on a higher and simpler plane. When the effort pays off, it can do so in a big way, since unconventional thinking often leads to big innovation.
What does all of this have to do with enterprise storage? Maybe everything.
While James Lau and David Hitz, co-founders of Network Appliance Inc., hardly fit the mold of "surfer dudes," the innovative idea they advanced of an information storage appliance has certainly earned them the mantle of employing thinking that was unbounded by the industry logic of the time.
After many years of laboring within the cubicle farms of various Silicon Valley companies, Lau and Hitz founded NetApp in 1992 on the simple premise that separating storage from servers and relocating it onto network-based appliance platforms would enable information to be managed more effectively, scaled more efficiently and shared by everyone who needed it. Their peers at the time must have thought that they were out of their minds.
While network attached storage has moved into the technology mainstream today, that wasn’t so in 1992. In fact until recently, conventional wisdom held that such a technology would lead to anarchy.
Sun Microsystems Inc., inventor of the NAS-enabling protocol Network File System (NFS) and long-time champion of the notion that "the network is the computer," didn’t release a NAS appliance of its own until this May. EMC Corp., long a NAS contrarian, argued for years that the storage appliance could never effectively compete with EMC’s high-dollar, high-performance, block-level storage arrays. Lau and Hitz must have smiled a knowing surfer’s smile a year ago, when EMC unveiled its first NAS products.
The success of NAS has turned the basic storage platform into a commodity. NAS is essentially a blending of a bunch of disk drives in a rack or cabinet, and a storage-optimized operating system kernel featuring NFS or Common Internet File System (CIFS—Microsoft’s answer to NFS) protocol support. With literally hundreds of NAS products on the market today, pressure is on developers at Network Appliance and other vendor shops to expand platform capabilities by continually adding features and functions. The competition is rarely spiteful however. NAS vendors demonstrate for the most part a courtesy toward competitors that’s unusual in this business.
You, the technology consumer, have benefited from NAS competition. Vendors compete on a more or less level playing field established by the IP networking and networked file system protocol standards that all NAS products must support. Regardless of the product you buy, you can feel confident that you’re getting good basic value in any NAS product. Additional capabilities added by vendors to further discriminate their products from the fray are, as a surfer might say, sweet.
Sadly, the experience of NAS has not translated successfully into other areas of storage networking technology development. Storage area networks—particularly the Fibre Channel variety—are a case in point.
The idea of a SAN—an intelligent, manageable and scalable storage utility arranged around a back-end network infrastructure—was advanced by several unconventional thinkers operating in different companies at about the same time in the mid-1990s.
What happened to SANs? Some might say that the pure break was polluted by commercial interests—the same thing that almost destroyed surfing in the early 1990s. Fibre Channel, not a network at all, was pressed into service as the SAN interconnect. It’s a job for which it was ill-equipped, but it was used simply because it was the fastest storage protocol available at the time. Vendors jumped on the hot technology and flooded the market with products that didn’t interoperate (and still don’t for the most part), despite the claims of many Fibre Channel advocates.
In place of true interoperability standards, the Fibre Channel SAN crowd has substituted periodic "plug fests" that resemble an overcrowded break.
The lack of a baseline of standards continues to make Fibre Channel SANs consumer unfriendly. Costs of SAN solutions vary radically from one vendor to the next and there is little reassurance that the basic value of a SAN will be realized from whichever so-called "solution" you decide to purchase.
Regardless of how appealing you find the thought of setting aside day-to-day storage administration challenges, few of us can engage in a life-long quest for the perfect storage network. The best you can do for now is to watch the swells, listen for the "Surfs Up!" warning and start paddling toward the shore. When you feel the wave launch you forward, pop up and ride it in. In surfing, as in storage networking, timing and judgment are critical.
Jon William Toigo is an independent consultant, and the author of 11 books, including Disaster Recovery Planning, 2nd Edition, and The Essential Guide to Application Service Providers, both from Prentice Hall. He can be reached at email@example.com.