Why ILM Doesn't Float My Boat

Information life-cycle management (ILM), recall, is the cradle-to-grave management of an enterprise's stored data--flat files, databases, everything. By adding management intelligence or a policy engine to storage, ILM can move data to cheaper media and keep a record of who accessed it and when. Ideally, ILM should reach beyond a single vendor's arrays, tape libraries or management suites for full-bore interoperability. But I'm getting a little ahead of myself here.

ILM is what's driving storage, whether to justify billions spent on acquisitions (as is the case with EMC) or to grab software's superior margins in a sector where hardware is largely commoditized. But what's being sold as ILM is also known as data life-cycle management, file life-cycle management or, as it was once known, hierarchical-storage management. In addition to the EMC-Legato-Documentum triumvirate, Veritas, StorageTek, Hewlett-Packard, IBM and Computer Associates, among others, are pushing some form of managing data life cycles. VARs who aren't with the ILM program should know what is going on, if nothing more than to protect themselves and understand how to position their experiences going forward.

Any so-called ILM products being sold today are little more than glorified backup-and-recovery or archiving tools. Such products have been snapped up by customers in sectors like health care and financial services, largely because of federal regulations, including HIPAA and the Sarbanes-Oxley Act, which require them to hang onto instant messages, e-mail and countless other document types.

This is no doubt a ripe patch for vendors, given that there are more than 20,000 such laws or regulations around the world protecting privacy and creating more transparency and accountability. Predictably, vendors have seized on the uptake of such products as proof for the demand of ILM wares.

Like most other marketing, this is equal parts wishful thinking and awareness building. The thinking is, "We may not have a rich suite of products for ILM processes, but we'll expose and educate and, perhaps, even plant the seed of demand."

But as some overpaid consultant recently intoned, "Hope is not a strategy." While all the economic temperature-taking allegedly shows more customers willing to spend more money, the "show-me" mentality and ROI pressures are no less present.

Take, for instance, a discussion at Storage Networking World in Phoenix. Customers are looking for a payback period of no more than three years. Today's ILM products can promise 30 percent to 50 percent annual cost improvements, but that's a high threshold for any software to achieve.

That explains the vendor backpedaling in recent months, as they've come to realize ILM's promises have gotten too big. Recently, Mark Lewis, an executive vice president at EMC, said it's going to be a while before the vendor delivers what he called "ILM Version 1.0." In the meantime, EMC will boost the capabilities of its VisualSAN suite. It also plans to begin testing a storage router that virtualizes data from other storage switch vendors, like Cisco, Brocade and McData.

That's a good start and is probably emblematic of how ILM concepts and practices will start to permeate the enterprise. Clearly, this isn't going to be some overnight flash-cut to the intelligent management and movement of stored data across multiple vendors' equipment. The real work will come in demonstrating tangible benefits and returns from any ILM purchase. Customers don't have to agitate for anything in this case; they can just continue to take a pass till ILM's economics improve. In the meantime, vendors can continue to beat the drum and educate and popularize ILM's precepts. But you can be sure they'll begin doing all that with a bit more pragmatism and emphasis on deliverable benefits.

Terry Sweeney ([email protected]) is a freelance writer and site editor of Storage Pipeline (www.storagepipeline.com), a VARBusiness sister site.