Yesterday it was announced that VMware and Amazon Web Services are partnering to provide vSphere’s hypervisor and toolsets on the AWS platform. Since this time there have been plenty of articles written questioning the motives of both parties involved and whether or not one of these two companies is going to regret this partnership. I invite you to read other perspectives on this and a few of them are listed here: Cloud Opinion, Enrico Signoretti, Frank Denemman (VMware), Jeff Barr (AWS) and there will be more.
What Does VMware Get Out of This?
VMware gets a quick, short term entry point into the clear number 1 public cloud vendor’s infrastructure. VMware has been trying to get vCloud Air off the ground and never really generated any lift. This deal, will allow VMware to run their solid hypervisor inside Amazon’s data centers and give users access to regions that would have been very costly for VMware to get into. AWS is likely providing the data center gear (Power, Cooling, Networking, Security) already and VMware just needs to manage the hypervisor, servers and overlays to their product such as vCenter. VMware seems to understand that customers are moving to the public cloud anyway, and want to have a viable solution ready for them before they make the decision to move. For VMware this is a way to hold on to customers who may be migrating to public cloud for a little while longer.
My Take: This is a good move for VMware for the short term, but probably also shortsighted.
What Does AWS Get Out of This?
Amazon Web Services Q2 revenue numbers reached 2.8 Billion and show no signs of slowing down. Cloud adoption is still increasing quarter after quarter and many customers are feeling pressure to move to an operating expense model where you pay for what you use, instead of depreciating equipment over three to five years regardless of how much they’re being utilized. Amazon has grabbed many of the early adopters but in order to continue this growth without a diminishing return, new revenue streams have to always be considered. Amazon should have little trouble bringing in new revenue from startups and development shops but many of the larger enterprises that have such a large investment in VMware and aren’t going to re-architect their applications for cloud at a very rapid pace. Sure, these customers will do some new development and architect it for cloud but what about all of the older “legacy” solutions that they have to maintain? This VMware partnership makes it very simple to “lift and shift” the workloads out of the physical data center and into one of AWS’s data centers. How do I know that these companies aren’t going to be rapidly refactoring their applications to move it to the cloud you ask? Because I work with companies every day that still have physical servers, mainframes, and other solutions that haven’t been able to migrate to a virtualized environment yet.
So, Amazon gets some more revenue by making it easy to move workloads into their data centers right? Many of us already know that just picking up a workload and sticking it in the public cloud will work, but usually becomes more expensive if the application isn’t re-architected to take advantage of scaling for on-demand resources. While VMware is busy selling AWS’s services, customers may start to see that these costs are just to expensive to deal with at which time Amazon can swoop in and show customers how they can eliminate costs by moving off the VMware solution and onto native AWS capabilities. I mean, the workloads are already in Amazon’s data center so this should be an easy transition.
My Take: Amazon is the clear winner in the long game. VMware will be selling Amazon’s product indirectly and getting users to adopt their the public cloud for AWS. Later Amazon will swallow up those workloads anyway.
What Does the Customer Get Out of This?
The good news in all of this is that the customer really wins. There is significant pressure on CIOs to move to “cloud” because it’s cheaper or to move to this Opex model for accounting purposes, or any of the other reasons businesses think cloud is important. This solution will make it easy for CIOs that have initiatives to move a percentage of their data centers to the cloud by 2018. The migration should be pretty seemless, with little downtime, no re-training of administrators, and no re-architecting needed by developers. The straight to AWS option is still there, the on-premises option is still there and now there is a new option which is VMware on AWS. Choice should always be good for a customer and this partnership gives them another one.
My Take: This additional choice will be an excuse for System Administrators to be complacent and to not learn new skills that may help them in their career. Instead of learning new skills, admins can be contented to leverage their existing skill sets. Besides this caveat though, customers will have an excellent choice for hitting their corporate mandates without putting the environment at risk by sloppily moving things to a new platform.
Its WAY too soon to tell whether any of this is going to really play out like I see it, but these are my predictions. Hopefully this is one of those deals where it works out for everyone involved.