Choosing a new primary storage solution can be a challenge for any organization. Aside from selecting a product set that will provide the capacity and performance you require -- a difficult task on its own -- you also need to consider how difficult the product will be to manage on a day-to-day basis.
Storage products' management software provides the lion's share of differentiating value. Whether you're talking about more complex features such as site-to-site replication or simple stuff such as being able to schedule snapshots or clone a volume for a test environment, all those features are essentially software features. The trouble is that very few potential customers ever get to run these features through their paces to get a feeling for how easy they are to manage and how much value they provide.
[ Managing backup infrastructure right is not so simple. InfoWorld's expert contributors show you how to get it right in this "Backup Infrastructure Deep Dive" PDF guide. ]
The result is that new primary storage owners invariably discover that something they just bought doesn't quite do what they thought it did. At best, this means that you may have to change your approach to a given problem -- perhaps by writing a script with vendor-provided API tools that will run on a schedule to perform a repetitive maintenance task instead of just being able to handle the scheduling through the management GUI. At worst, you may have just spent a pretty penny on a product that simply doesn't do what you thought it did or realized afterward that there was an expensive license you needed and weren't sold. That's a position nobody likes to find themselves in and is challenging to completely avoid.
The traditional way for a new customer to learn this sort of information from a prospective vendor is through a demonstration. The vendor might invite you to a lab to get your hands on it or, if you're lucky, might come to your site and set it up for you and let you kick it around the block for a while. That's great, but how much can you really learn in a few hours while you have a vendor looking over your shoulder pointing out all of the strengths of its product?
In a time when I can freely download all three of the most popular virtualization hypervisors, install them, and run them for a month or so before I make a decision or pay a cent for them, you'd think there would be a better way to evaluate storage. Unfortunately, because SANs are predominately hardware solutions, there isn't. You can't just download a SAN. Or can you?
Software simulator can help -- but just existing customers
Several major storage vendors, including NetApp and EMC, have offered software-only simulators of their products for a few years. A colleague pointed out this recent blog by a NetApp systems engineer that talks up NetApp's Data OnTap Simulator. Data OnTap is the management software that most NetApp storage devices run, and this simulator is essentially a full-featured, hardware-independent version of that same software.
With a bit of know-how, you can get the simulator set up and fully emulate all of the software functionality of a real NetApp storage system. I don't believe the simulator image includes NetApp's Operations Manager -- a centralized management console that is usually a key feature of any significant NetApp deployment -- but it does let you work directly with the SAN's interface and kick it around to your heart's content. As long as you're not trying to gauge performance or use any features that require specialized hardware (FibreChannel, for example), it is a great learning tool.
Before you get too excited, though, this simulator is only available to existing NetApp customers. Which makes some sense, given that it is partially supported and perhaps not particularly easy to get running if you're not familiar with Linux. The simulator's primary goal is to let an existing customer test new software or test advanced software integration features outside of production. That's excellent -- NetApp should be congratulated for spending the extra effort to provide their customers with such a potentially useful tool.
However, that doesn't help you if you're not a customer yet.
Storage vendors need to let customers truly try before they buy
As someone who is often in the position of helping a client choose what storage option will work best for them, I would positively love to be able to fire up a collection of virtual machines that could emulate the software features of all of the major SAN vendors' flagship products: "Are snapshots a key interest of yours? Great -- here's how manage them on these six different products."
Technically, it should not be hard to do, and in fact, this is the way that much of this software is developed and tested by the manufacturers themselves. So, if you're a storage vendor and truly proud of your product and believe it competes well, why not make the simulator available to everyone?
The answer, of course, is that even the best and most honest of these vendors have at least a few warts they'd rather you didn't notice by applying some sales makeup. Controlling and guiding the sales process is many times more important than having a fully educated customer. And, to be perfectly blunt, some storage vendors would promptly be out of business if they had fully educated customers.
However, I hope that increased competition and pressure from prospective customers makes this concept a reality. It would be an excellent way to foster more competition and would force storage vendors to work that much harder to develop new and unique storage management features within their products. Sadly, that's also precisely the same reason it may never happen.
Perhaps the first step is that customers -- who pay the bills, after all -- start demanding simulator access. That increases the chances you'll actually get it.
This article, "Can you avoid a big mistake when buying primary storage?," was originally published at InfoWorld.com. Read more of Matt Prigge's Information Overload blog and follow the latest developments in storage and data management at InfoWorld.com.