Ours is an industry that tends to overuse and abuse words more than most.
We use them too frequently, and too imprecisely to the point where they lose their ability to convey significant meaning.
And, on the list of maligned IT words, I hold a special place for the word "solution".
In our industry, it's applied to so many things, it has come to mean almost nothing.
I've come to call it the "S" word -- to let people know I consider it impolite and inappropriate to use in public unless you're willing to take the time to convey what you might mean by it.
EMC Is Guilty Of This As Well
Over the years, we too have used the word "solution" casually to describe all sorts of things. We didn't have a clear model and taxonomy of solutions, each with their associated characteristics and attributes.
But we're starting to get better -- much better in my view.
First, we've started to organize internally around solutions. One of the results is that no longer is it possible to just slap the "S" word on whatever comes along without having it mean something very precise.
When we use the word "solution", we want it to mean something very tangible and specific, and communicate clear expectations to our customers that they can depend on.
And some of this new approach is becoming externally visible.
The Recent VMware Solutions
It was a busy week, but in the flurry of announcements from EMC and other vendors at VMworld in Cannes, we announced our first range of solutions for VMware environments that were targeted at particular applications.
Unlike other "solutions", these actually mean something very specific, and somewhat unique, in my mind. And, hopefully, you'll see more of this sort of "solution" from EMC and others in the future.
Let's say you wanted to run something like, Exchange, in a VMware environment. Or Oracle, or SAP, or SQLserver -- pick your favorite.
You'd want to know what kind of performance to expect, wouldn't you? You'd like to know how to best back it up, or replicate it, or archive its information, or perhaps secure it.
You'd really appreciate a "cookbook" of sorts that told you how to assemble the ingredients to get the results you wanted the first time, and without a lot of fuss. And, ideally, you'd like to be able to call a single vendor if the results didn't quite turn out as you'd expected.
That's the idea with EMC's VMware solutions.
We start with an idealized use case, say 5000 Exchange users. And then we pick out the best components for the job: servers, storage, network, backup, and so on. Sure, there's a lot of EMC product in the solutions we produce, but there's good representation from other vendors as well.
We assemble and qualify all the pieces. We characterize the performance of the environment in production, backing up, replicating, archiving, and so on. We document everything we learn, and provide step-by-step instructions on how to assemble it. We suggest good ways to migrate from old to new.
And, if there's a problem along the way, we're happy to support it.
And the best part is that we make it available not only to EMC employees, but to our partners and customers -- at no charge.
Why Is This Important?
The real expensive part of IT these days is labor. Whether it's your own staff's time, or you're paying a vendor or contractor to do something, labor is becoming a dominant component in IT.
Not only that, time is money. If you're waiting for the benefits of, say, running your new Exchange environment virtualized, you'd like to get there in six days rather than six months. And if you've never done a particular kind of project before, you're not really sure what you might find along the way.
Simply put, I think the value of this new breed of EMC solution is simple -- it gets you to good sooner, and with less risk. Less possibility for expensive mistakes. Documented results at the end that you can count on.
I think that once people fully understand what we're doing here, it's going to be a very popular concept.
But My Environment Isn't Like That!
So, maybe you take a look at one of these VMware solutions we've created, and you say to yourself that you use a different flavor of server, or backup, or whatever.
Fine -- the EMC solution approach still has value. You can look at our reference architecture and documentation as a starting point, and get a nice leg up on your project just by looking at what we've done, and why we did it the way we did. The more you deviate from the template, the more you're back to a traditional integration and support model, but the choice is yours.
What Does This Mean?
I think that all IT infrastructure vendors realize that they need to think in terms of customer use cases, rather than individual products that they sell. And, while it's great to have skilled people who really understand solutions in different environments (we'll always need those!), that can be expensive.
I really like the idea of taking what we've learned in popular application environments, documenting it, and making it available in an easy, useful form that delivers tangible results, time after time.
If you think about it, EMC has done this before in a different context.
When SANs were new, and interoperability a nightmare, we invested heavily in EMC eLab, which ended up being the de-facto reference in the industry of what worked, and -- more importantly -- what was known not to work.
It was a very expensive investment at the time, but if you were close to the action back then, you'd probably agree that it moved the industry forward in a big way.
In some sense, our investment in solutions is like the next generation of eLab -- we're bringing together diverse technology components from different vendors, testing and qualifying the bejeezus out of them in real-world use cases, characterizing performance in a variety of modes, documenting everything we've learned, and making it accessible -- at no charge -- to people we're doing business with.
Make no mistake, it's expensive to do this sort of "solutioneering" -- very expensive. I don't think many IT vendors (or other storage players) are up to investing in this area to the degree we are. I'm sure that they'll eventually claim something similar (they all tend to offer a "me too" after EMC does something), but I'm skeptical that the results will be the same.
And, although the first results are pretty cool in my opinion, there's a lot more work to do. I wish I could post the rich technical details here, but I can't. It's great, useful, meaty, valuable reading.
Including a few eye-opening findings.
I'm glad we're doing this. I think -- over time -- it'll make a big difference to everyone.
I suppose one of the reasons I like working at EMC is that we tend to make the big investments in areas we really care about - including making our customers successful with VMware.
Comments