In-Depth

Winning the Little Battles to Demystify SMI-S

What, exactly, is SMI-S anyway? We may finally be able to find out.

• By Jon William Toigo
• 05/06/2004

Just back from Storage Networking World, I must say that I’m getting tired of hearing different definitions of SMI-S depending on which vendor I ask. Is it BlueFin, a storage device discovery technique, blended with that Common Information Model/Web-Based Enterprise Management (CIM/WBEM) stuff SNIA was buzzing about a couple of years ago? Is it a panacea strategy for configuring heterogeneous storage gear from a single console? Is it a floor wax that's also a desert topping?

The question became an issue earlier this year when we quoted a very knowledgeable storage veteran in a story on storage management. The source told me, “SMI-S is just another API swap … worthwhile only for as long as vendors consider it to be in their best interests to cooperate. Once a vendor gets the upper hand and decides to go his own way, SMI-S will be pretty much worthless.”

Those were fighting words to members of the key SMI-S promulgator, SNIA, an organization whose existence is difficult to justify if SMI-S doesn’t come to fruition. I received a hostile e-mail stating that anyone who viewed the burgeoning specification as “an API swap” clearly knew nothing about it. That prompted my search for a cogent description of the technology and its business benefits.

Alas, my efforts ran into obstacle after obstacle. In truth, SNIA has done an awful job of facilitating an understanding of SMI-S among end users. I can’t begin to count the number of SNIA-driven educational sessions that I have attended at various storage events, each promising to define SMI-S in terms and language that everyone could understand. These sessions, however, were always the same: a 45-to-60 minute lecture on object oriented programming and metamodels. Looking around the room at the glazed-over eyes of other attendees, it was clear that the SNIA speakers were missing their targets.

At SNW in Orlando last fall, I challenged every SNIA-ite who would listen to do something about it. That explains my delight when I received an announcement via e-mail a couple of weeks ago stating that SNIA was planning a Webcast to demystify SMI-S. It seemed a small victory.

My jaw dropped, however, when I saw that the organization was planning to charge attendees $150 for the privilege of viewing the e-show. Inquiries made to the media house handling the event revealed that it was not the media house, but SNIA itself, that had attached a price tag to the presentation. Apparently, someone thought the industry association’s first foray into the world of quasi-standards-making should be a pay-per-view extravaganza, like world wrestling smackdowns, or boxing from Caesar’s Palace, or lingerie model football.

Someone missed the point, I argued to the SNIA director at this month’s SNW. Vendors were already complaining, I explained, about the costs for engineering SMI-S providers into their gear. Some viewed the technology as the latest in a litany of engineering requirements that they had been compelled to meet in their gear without any guarantee that consumers would pay for them. Said one vendor, “We spent money building JINI support into our products, then other specifications, and now we are told to spend scarce dollars to add an SMI-S provisioner to our gear. It’s not like customers are lining up around the corner to buy products that are SMI-S enabled. They sure aren’t showing any willingness that we can see to spend a little extra for SMI-S functionality.”

The fellow was reiterating the rationale that many vendors had proffered recently to explain why they had dropped their participation in the SMI-S specification development process. Each of these vendors said that SNIA had done such a poor job in generating grassroots support for the technology that they couldn’t define a compelling business reason to justify the expense of supporting the effort.

To be sure, some of these comments may have represented sour grapes on the part of the vendors making them. That aside, however, it was clear that SMI-S lacked a clearly articulated and compelling business case capable of cultivating consumer momentum behind the spec.

SMI-S remains a mystery to many people, which is why it was enigmatic in the extreme to charge money to view a Webcast that SNIA-ites promised would make the business case from the consumer perspective. Somebody wasn't paying attention in Marketing 101 class.

Anyway, I made this point to the SNIA director and was surprised and delighted to be cornered the following day by some SNIA Board Members who informed me that the charges for the Webcast have now been dropped. Another little victory.

SMI-S may be the greatest thing since sliced bread, or it may be all potatoes with no meat. But SNIA promises that we will at least know how to read and understand the menu after the Webcast, which you can register to view online at the following URL: http://www.iian.ibeam.com/events/penn001/022604a_by/. The archive for the event is available for replay through mid-June.

We’d be interested in hearing from you following the event to learn whether it succeeded in motivating you to buy SMI-S products: [email protected]