Lock-Outs for Independents

Is there a role for the independent integrator in the well-demarcated world of storage vendor relationships?

Interestingly, this past week has seen an uptick in e-mail from independent storage integrators complaining about a trend they say is locking them out of their own accounts and limiting the kinds of solutions they can put together for their customers. Here's a sampling:

“It seems that if a VAR like us is not completely aligned with certain vendors and their products, including OEM products, the VAR stands a big chance of losing out on [an] entire deal or solution sale. I continue to see this as we go out and pitch our solutions. The biggest one I … ran into recently is in the SAN market. We are very big on QLogic products mainly for the price and performance and feel and see that customer[s] like it as well.

"However if you go into an account with EMC, Network Appliance or StorageTek, they are not willing to support their product in a QLogic environment. Only when push comes to shove will they agree, and only if it means a sale to them. From where I sit, this looks like deep, very deep, OEM agreements and I would consider this a proprietary configuration [from vendors who not long ago were the champions of open systems].”

Are comments like this just sour grapes from an integrator who isn’t on the right party invitation list, or might this situation be testimony to the increasing commoditization of storage products that will eventually relegate the storage integrator/reseller to the history books?

While contemplating this issue, I found myself recalling a bit of my own history. After about 10 years of corporate tech employment, I worked for a couple of integrators through the mid 1990s, first in the Washington, DC beltway area, then in Florida. What sticks with me about the experience is the nobility of the integrator’s mission: we sought to cobble together, from a variety of product sets, the right combination of gear and software to meet the client’s business computing requirement.

To be sure, we did not have a full and detailed knowledge of every product produced by an industry well known for launching new stuff with the speed and alacrity of happy rabbits. But we did the best we could to find “best-of-breed” brands (a categorization based largely on product failure rates diligently recorded in deployed solutions) and to vet everything we put together through rigorous testing. As a result, we rarely, if ever, heard a client complain that they were tired of being “the point of integration” for miscreant technologies, an objection I commonly hear in the FC SAN world today.

The reader’s objection goes, in part, to this point. His firm has found a technology vendor in QLogic that consistently delivers value in connection with his storage solutioneering efforts on behalf of clients. He is confounded by the fact that other vendors will not “certify” or “warranty” or “support” their products when used in conjunction with a QLogic product, not because of some demonstrable problem they have discovered regarding the interoperability or solvency of the solution, but because of OEM agreements they have with other vendors.

This practice is as old as FC SANs, of course. The endemic interoperability issues in FC SANs hearken back about five years to the original rollout of FC technology into the mainstream. Vendors created little cliques among themselves early on, co-marketing vendor X’s switch with vendor Y’s array. EMC had its posse, and so did Sun and IBM; HP had a lot of difficulty with commitment to any single suitor. This was good for newcomers to the market (switch makers, predominantly): sort of a Big Brother thing. It worked for the big iron manufacturers, too, because they only needed to test their gear with one vendor rather than all vendors.

To hear a representative from EMC tell it, the folks in Hopkinton believed that such relationships were key to their “we service what we sell” philosophy. Said the rep, EMC “owned” any solution that it deployed, servicing it after the sale as part of lucrative maintenance agreements. He believed that it would be problematic for any array vendor to allow its customers or independent integrators to cobble together “untested” combinations of its branded arrays with third party switches and HBAs. When the solutions failed, for whatever reason, it would always reflect badly on the name-brand vendor.

On the other hand, vendors like Dell have sought to “productize” storage solutions as an expedient to sales. Today, the Web retailer ships “SAN in a box” solutions that combine the products of multiple OEMs in “replicable configurations.” One could argue that productization (another way of saying commoditization) is the inevitable result of technological maturation. Microsoft’s Enterprise Storage Group has taken this idea to heart as it seeks to build the technology for deploying and managing SANs directly into its operating system. Someday soon you may be able to download a bulletin from Redmond that spells out exactly what hardware will work with its server functionality to build a SAN.

So, there are two explanations for how we find ourselves in the predicament cited by the integrator quoted at the outset. What neither explanation provides, however, is an answer to the question: is there a role for the independent integrator in such a well-demarcated world of storage vendor relationships, or should they close their doors and seek gainful employ elsewhere?

Truth be told, there is no one-size-fits-all storage platform. Despite efforts to dumb down FC SAN technology so that a drooling idiot can deploy one, this goal continues to be confounded on many levels.

Stating the obvious first, different applications require different storage solutions, which makes every SAN deployment, however “out of the box” that its vendor may claim that it is, a custom project. While there is nothing inherently wrong with a Dell (or a QLogic, for that matter) plug-and-play SAN, it isn’t right for everybody—or every app. Independent integrators evaluate application requirements and select the best topology and components to meet them.

If array vendors will not “certify” or “warrant” their gear to work with other vendors outside the clique, it is up to the integrator to prove to the client that the recommended solution will deliver represented value. This can be an uphill battle, of course, depending on the client’s desire for vendor-based maintenance, but it is also one that can open the door to a potentially more lucrative business for the integrator: post-sale solution support.

More good news: in my experience, any contest between vendor and integrator in just about any client environment (save for the Fortune 500) usually works to the integrator’s advantage. As a local solution provider, the integrator brings a human face (and one throat to choke) to the value proposition, versus an anonymous vendor brand. More storage has been sold by local integrators, who enjoy more face time with prospective buyers and do the bulk of the knowledge transfer and education that are prerequisite to the product sale, than all the direct sales staff of all leading vendors combined.

Bottom line: as Sun Tzu once wrote, you need to use your opponent’s strengths against him. Leading storage vendors may be marketing machines that can churn out SAN-in-a-box solutions the way that McDonald’s makes its milk shakes. But just as no one prefers milkshake-from-a-mix to the handmade-by-the-ice-cream-parlor-guy refreshments, no one really prefers one-size-fits-all storage to hand-tooled solutions tailored to application needs.

Your comments are welcome: jtoigo@intnet.net.

About the Author

Jon William Toigo is chairman of The Data Management Institute, the CEO of data management consulting and research firm Toigo Partners International, as well as a contributing editor to Enterprise Systems and its Storage Strategies columnist. Mr. Toigo is the author of 14 books, including Disaster Recovery Planning, 3rd Edition, and The Holy Grail of Network Storage Management, both from Prentice Hall.