A few weeks ago, I was with a customer -- a large retailer in the Midwest -- working to close a deal before the end of the quarter. As the customer pointed out, last year this same deal would have generated about three times what this bid was worth.
We did the deal, but this customer's comments portend where the industry is heading and what any storage vendor looking to stay in the game must address. The customer said to me, "Look, you and all of your competitors are offering me basically the same stuff. Same capacity, same capability. What I want is a partner that can cut through the noise of open vs. closed, FICON vs. IP, SAN vs. NAS. I have storage from all of you vendors, and I'm sick of all you guys looking at me like I'm a bank with these one-off transactions. So how do you plan to address my future needs?"
Tough question, but the right one to ask. The answer for all storage vendors, in my view, is quickly morphing toward two differentiators -- technology innovation and integrated solutions.
The old days of just add more storage are long gone. In fact, it was just that mindset that created the islands of automation that customers are now trying to overcome. Recent studies have found that the initial purchase price for storage hardware accounts for only about 10 percent of the total cost of ownership.
While more and more storage is being moved to the network in order to improve systems performance, scalability and data availability, the pressure is mounting on the storage administrator, acting as a kind of "Wizard of Storage" behind the curtain, trying to manage all of these applications and data while masking the complexity from the outside world.
What customers need is innovative technology that will virtualize the hardware while delivering a single file system. Imagine all storage managed as one resource, and the ability to share files seamlessly among servers in a SAN. These two innovations will allow customers to exploit the existing, underutilized capacity resident in their storage hardware. Only then will customers begin to see the real TCO and ROI benefits that we've all been talking about over the past few years. If you haven't imagined this brave new data world -- ask your customers, they certainly have.
The second piece of the data storage puzzle is integrated solutions.
Customers are no longer willing to be "quilt makers" for our storage piece parts. It's too hard and it's too expensive. In trying economic times, it's imperative that we provide -- both on our own, and as an industry -- open networking solutions that take the customer out of the middle of our compatibility problems. This is why adherence to Storage Networking Industry Association (SNIA) standards, standardized application program interfaces (APIs) and support for Storage Solution Forum initiatives are becoming must-haves for all customer discussions. Single-vendor plays, with promises of backward compatibility and reverse engineering, may suit the best interests of that vendor, but not the customer.
This is why it is imperative that initiatives like Project eLiza, IBM's blueprint for making systems more autonomic, be embraced by the storage community. Self-managing, self-configuring and self-healing storage systems are critical as storage moves from the back room to the center of the IT infrastructure. The application servers and storage servers are "six-nine" devices, and the storage infrastructure must maintain the same reliability level as it moves to the heart of the network.
For most customers, the scope and complexity of the storage infrastructure is fast outgrowing their ability to manage it manually. Customers need storage systems that will care for themselves.
I came away from my recent conversation with that Midwest customer having confirmed a very stark view of the future of the storage business. As a vendor, you'd better innovate and integrate -- and create enough added value to justify a bit of margin in your bids. Competing on price alone is not a sustainable business model, but customers will pay a premium if you can deliver a compelling value proposition -- and a roadmap -- to the future of storage.
IBM laid out a detailed storage software roadmap in late April, a direction aimed at adding new levels of intelligence to storage networks. The roadmap targets three areas of innovation: a Linux-based virtualization engine, IBM's Storage Tank file system, and storage networking designs based on mixed-vendor interoperability guidelines set forth by SNIA. The virtualization engine, which will enable administrators to view and access a common pool of storage on a network, and Storage Tank, the first file system optimized for storage networks, are both slated for introduction in 2003.
Customers need not wait to start migrating to this type of new model of storage technology. They can get started today -- by moving away from the direct-attached world and attaching storage to the network. As new software capabilities become available, customers will be able to incorporate them into their existing storage networks.
Copyright 2002, IBM Corporation.
This was first published in June 2002