In this blog I look beyond the obvious and try to find out why people and companies do what they do - and what it means for the rest of us. 
And ESG's soft spoken virtualization guy Mark Bowker is in the middle of all of it. Of course, he's in Cannes and I'm freezing my rump off in Boston, but someone has to do it.
The NY Times article, "Microsoft, Citrix Join Forces Against VMware" has Bowker correctly calling the game as it is, effectively saying this is really about MS and Citrix and putting their stuff together because individually they fall way short where VMware rules. I personally am not yet convinced it matters – at least as far as the data center server consolidation space, I think VMware owns it.
The fight, however, isn't there anyhow. VDI – desktop virtualization is WAY more important to both MS or Citrix – the sheer volumes make it an outrageous market opportunity. You can think of it like this – what if every desktop in every company on the planet suddenly DOUBLED. That's the opportunity for MS et al. They literally have just doubled their total available market opportunity with VDI.
Mark is quoted in this CNN Money article saying VDI ain't gonna happen as fast as the hype – even though I personally believe it is one of those rare no-brainer shifts that has to happen - due to the big back-end infrastructure investment. He's right, of course, but to me it's completely inevitable. MS, Citrix, VMware etc. have huge stakes – but think about the ramifications elsewhere? At first, everyone is happy because just like the VMware explosion, it will create massive net new churn – people will buy tons of storage and servers and licenses etc. which will make everyone happy for a while. Eventually, however, it will mean Intell and AMD will suck pond water, and the suppliers of all the desktop components will cry me a river. Who makes all the desktop disk drives these days? They will be bumming.
He was also in thisForbes article yesterday talking about VMware and the "Cloud". VMware calls thier high level construct the "virtual data center operating system", which is wrong – it should be just Data Center OS. Why? Because 5 letters is too long, there is no need to use the word virtual anymore, and people will actually draw the appropriate understanding of what a DCOS is. Plus, I came up with DCOS years ago at a Storage Decisions conference and damn if their story ain't 98% the exact same as mine. They owe me at least a billion bucks as far as I figure.
He who wins the DCOS wins the power of the data center moving forward. The stakes are enormous. All who sit beneath it will have lost all their control value, and will be completely (finally) commoditized. Do you know what's inside your laptop? CPU's, memory, I/O and other widgets. Do you control any of them? Nope. Who controls it all – the OS of course. Same thing with the DCOS. Big stakes.
In case you haven't figured it out yet, within 12 months EVERY server shipped will come (one way or another) with a hypervisor included. Within 24 months EVERY computer shipped will come with a free hypervisor. Hypervisors are gonna be just like BIOS. What does that mean? It means that machine/server/processor or whatever you want to call it – Virualization ain't a "maybe" anymore – it's done. What you do with that hypervisor will vary – some will do zip, others will do lots – but the fact that it will automatically just be there means it will be used eventually. That fact will change everything my friends.
It is a fait accompli – therefore, when virtual machines are absolutely commonplace, what might that mean for you, me, and the world? Disruptive force? I'll say.
Finally, on this fine day, did you see NetApp's latest announcement? Brilliant, or at least, very clever. I don't give my friends out there marketing kudo's very often, I love this one. They guarantee that you will save at least 35% of their existing storage capacity when customers go through a server virtualization move – on EMC, HDS, IBM, and HP. It's superb messaging. They are saying – "oh, you run their stuff, no problem, buy our V-series head and stick it in front of your storage boxes (and behind your virtual servers) and we'll give you snapshots, de-duplication, etc. and give you 35% or more capacity."
Why is this brilliant? Because A: these are non-NetApp accounts, and in one fell swoop they not only stuff themselves into those accounts, they also kill most new sales those others might have made since people are getting used to thinking they have to buy MORE storage when they go to VMware, and B: because they just inserted themselves as the new storage VALUE provider and relegated the boxes on the backend to expensive dumb capacity. Who do you think will get the next deal if that works? Genius. Oh yeah, plus they get net new revenue for both the V-series AND professional services – as per the rules.
The rules: "Customers anywhere in the world who purchase a new V-Series system for their virtual desktop and server environment, including Citrix XenServer, VMware ESX, and Microsoft Windows Server Hyper-V™, can take advantage of this program. If customers don't improve their storage capacity on competitive storage systems by at least 35% with NetApp V-Series and deduplication (and after following established best practices implemented by NetApp Professional Services), NetApp will provide the additional storage capacity needed to meet customers’ shortfall, valued up to 35% of the original controller purchased, at no additional charge*. Qualifying service delivery is available from NetApp Global Services or participating Authorized Professional Service Partners in networked storage for virtualized platforms."
I have to admit, I expect this kind of "punch you in the head" marketing out of EMC, but probably wouldn't have put NetApp on the short list of companies to take such a ballsy position. Nicely done.
The downside? Not much that I can see. They make the rules and if they don't come through for some reason, they just have to give the customer capacity until it works. Disk is cheap, in case you haven't noticed. Their worst case scenario is they give the customer capacity at whatever it is they charge, up to 35% of the original cost of the V-filer – which probably represents no more than a 20% discount. They really can't lose. Will people do it? I don't know, but it's a damn good way to get some attention.
Related posts:
blogs
Probably should do the math on this NetApp “guarantee” before getting all fuzzy about it.
If you have 300TB of existing storage, NetApp has to give you back a little less than 100TB. This is only a win for the customer if the cost of the vSeries gateway is less than the cost of another 100TB of disk – which is cheap, as you say.
And they should also be sure to consider the operational and infrastructure costs – not just the new hardware, but the added cabling infrastructure to add the NetApp “bump in the wire” to the existing FC fabric, downtime required for each connected server, reconfiguration/relocation/compression time, etc.
Oh, and if they FAIL, they have to give you ANOTHER box with the missing capacity.
I’ll agree – good marketing, but not necessarily a Good Value.
——–Coming from a position of Anarchy, you should know why I love this! It is exactly that – great marketing. As I said, I don’t know if anyone is going to do it or not – but the buzz and ballsyness (?) of the move puts them into a no lose situation, and at the very least has you and I yapping about it, which in turn makes some others that might not have paid attention do so.
So, for arguments sake, lets say I completely agree with your logic and that when ALL the elements come out the customer finds out that its marginal or no real value in the guarantee itself – but who cares? By then the mission is accomplished – they have shoved themselves into a potential deal where they probably wouldn’t even have known about it previously.
You and your brilliant company know as well as anyone that you can’t win a battle you don’t engage in. Second, we all know that if we are the odd man out, what we ideally need is A: the opportunity to engage, and B: the opportunity to show our value. If this move gives them the chance to do both, there will be some – regardless of the “guarantee” – who will see value in what they do. I know you don’t want that to happen – and I sure don’t blame you, but it’s a fact. The same holds true for you – in the accounts where you are not a significant player, what you want is the opportunity to battle – because you know that if you get the chance to show your value and the customer looks like X, you have the best stuff. If you don’t get to play, it doesn’t matter. If you do, and your right, then you win more than you lose. It is a strategy that has served you well my friend. Don’t begrudge others for leveraging the same strategy.
So, allow me to restate – I love the play from a marketing perspective. I think it accomplishes everything you can want and more – it generates buzz (we are promoting it, are we not?), it is interesting and compelling at the high level (with enough buzz and coverage, there will be users who see it, and the message is clear and universal – “we’ll guarantee to save you a pile of dough”, will make a big percentage of that audience go to the next step – which means it has done its job).
I was born at EMC, so fret not, I never confuse perception with reality. Reality doesn’t matter.
Good comment – thanks. ——-Steve
Like I said, I admire the marketing hutzpah – truly I do.
But rather than ME harp the opposing viewpoint, let’s include a customer’s perception:
http://storagebod.typepad.com/storagebods_blog/2009/02/v-is-for-value.html
As Martin points out, you can probably recover MORE than 35% just by wisely leveraging the thin provisioning and snap capabilities of your current platform.
And avoid having to migrate to a totally new environment to boot…
In fact, if NetApp’s gambit causes people to look closer at their current wasteage, it could well be bigger a benefit to their competitors than to themselves.
Always fun chatting with you, Steve…
- Barry
—–My point precisely oh master of mayhem. The customer is making thier point almost – they KNOW they can cut consumption in half, because they KNOW (as you do) that 50 years of compound, almost comical, innefeciency when it comes to storage can be undone fairly easily – that’s what makes it so good from a marketing perspective. It is why they can appear so aggressive – because they KNOW.
It’s sort of like being a good analyst. 95% of the time, a good analyst states what is obvious – emphatically. If you scream “killing baby whales is bad” really loudly, people think – wow, that person is passionate and clearly smart, as I also believe killing baby whales is bad. Duh.
Your final point on the “backfiring” is ALMOST legit. If and when NetApp is successful at making this an issue in an account – they very well could end up losing the end deal to you, or someone else. That’s not the point however – they aren’t opening up this can of worms in their OWN accounts – they are injecting themselves into you, IBM, HP, HDS, etc. accounts – thus why I contend that there is no real downside for them, thus why I contend it is masterful.
Want to know the best part? NetApp, as a general rule, sucks at marketing – at least for the last 15 years or so. I mean that with all due respect, mind you. It sure hasn’t held them back, so you can understand why it hasn’t normally been on the top of the “must fix” list. I should also point out that pretty much every marketing person I have ever dealt with at NetApp has been good – so it is a bit weird. Without knowing who’s brain figured this play out, it is a breath of fresh air and Jay Kidd should be commended. Having said that, I’m have tried but just can’t get on board with their new branding, but that’s another story.
In our business, as you know, most of the time good marketing hides inferior products, markets, or other warts. In this case, it’s just good all around.
And look, we’re still talking about it!
—–Steve
[...] they may miss from their Windows system. VMware Fusion is easy to use, and doesn’t require any additional hardware. You just install the software, then access the Windows operating system from your Mac. There [...]