At issue: SMI-S is gaining acceptance among users, but until vendors offer products with complete capabilities, full-scale adoption could be a way off.
Just one year ago, the Distributed Management Task Force Inc. (DMTF) announced that the Storage Networking Industry Association (SNIA) had certified the first products under the Storage Management Initiative Specification (SMI-S). SMI-S is based on the DMTF's Common Information Model (CIM) and Web Based Enterprise Management (WBEM) standards. SMI-S promised to allow end users to implement standardized management interfaces that would simplify storage management.
In the meantime, SMI-S has gained momentum. A SNIA program has tested more than 200 vendor hardware products from 17 companies in the last year. Perhaps more importantly, SMI-S has set higher expectations among the user community.
"I think that customers who understand the goals of SMI-S are really in favor of it and want it to succeed and will buy products that are headed in that direction," says Randy Kerns, senior partner at The Evaluator Group, in Greenwood, Colo. However, he says, the real issue is the fact that all the goals of SMI-S are not yet realized, and that many functions won't be for a while.
"Customers want tools to get their jobs done and that integrate to make life easier," he notes. The risks, he says, are impatience -- customers need solutions and will go with what works. They won't wait, especially if is only to get a product in which the SMI-S capabilities are still incomplete. "So, if a customer deploys a solution that is not SMI-S compliant for what is needed, it will be some time before they make a change -- thus postponing the overall success," he adds.
A recent article written by Geoff Hough, director of product marketing at 3PAR, paints a similar picture of SMI-S falling short of expectations. Like Kerns, Hough says that most customers and vendors support the concept of SMI-S. However, early adopters are being forced to make big investments, even though they will only be able to get partial SMI-S benefits. Admittedly, he adds, "for them, the benefits outweigh any issues they must actively work on to resolve interoperability and functionality." By contrast, a group he terms "midstream adopters" are putting off adoption until SMI-S demonstrates more commercial and technical maturity. Finally, of course, late adopters are staying on the fence -– waiting for the day when SMI-S finally lives up to its ideal.
Perhaps the harshest words regarding SMI-S compliance come from Steve Duplessie of the Enterprise Strategy Group. He cites a recent ESG survey which shows that users are determined to purchase only standards-based products, primarily because they are concerned about runaway complexity. Duplessie says users shouldn't "settle" and instead should only give business to companies that have put their full support behind SMI-S. However, Duplessie questions how deep and how strong that vendor commitment really is.
"In fact, vendors like EMC, with so many different products, would benefit [from SMI-S] as much as users," he adds.
For more information:
Are SATA drives ready for the enterprise?
Topics: SAN planning and design
Cheap SANs -- Hype or hot?
About the author: Alan Earls is a freelance writer in Franklin, Mass.
