As you may or not be aware, today VMware unveiled their newly minted partnership with AWS.There’s been one article by Frank Denneman that takes a closer look at this partnership. There has also been an oped? piece published by our favorite cloud parody account, @cloud_opinion.
Señor _Opinion states in his article that AWS has blinked, that they are feeling pressure from Azure and that they have basically had a knee jerk reaction and partnered with VMware in order to address the enterprise arena, tactically. I agree with some of his/her opinions, especially about Azure having a much better story in the enterprise than AWS, but I don’t agree with all of it. Case in point:
- “Customers that buy into this partnership will end up wasting money and time and will not be moving to Cloud, while paying the Cloud premium.
I don’t agree. We don’t have nearly enough information of what this looks like to make this sort of statement. Could this happen? Sure. If this turns out to be vCloud Air, but in AWS datacenters, then this will absolutely fail and be a wasted effort for customers. Which, if you read Frank’s article, it looks like it may be just that, vCloud Air in AWS datacenters; more on that in a minute.
- “They display tremendous discipline in staying focused on solving customer pain. Nothing about this announcement shows that discipline.”
Again, disagree. AWS is still focused on customer pain, and are directing some of that focus towards a particular market segment; enterprise. Many enterprises are already using public cloud, and the amount varies, but it’s almost all new development. New development is great, but what about everything else? In a magic land where all the Fortune xxx companies rewrite all their apps to take full advantage of public cloud, then their isn’t any need for VMware Cloud on AWS, but [most of us] we don’t live in said land. Enterprises are still looking for new ways to scale their applications that they’ve been using for years without always having to buy new hardware and go through 3-5 year refresh cycles with <insert your favorite vendor here>.
Now, Dhr. Denneman’s article outlines, from a high level, what VMware Cloud on AWS looks like. Leveraging bare metal servers inside of AWS datacenters, VMware will deploy a full SDDC stack (Cloud Foundations) on top of that, allowing the ability to move workloads private -> public, public -> private and public -> public. Outside of the fact that this is sitting in an AWS facility, it’s basically vCloud Air. We all know how well that worked (it didn’t). Hold up, here’s what is intriguing to me;
- “Another strength it the ability to pair current workloads with the advanced feature set of AWS. As a result, IT teams will be able to extend their skill set discovering the vast catalog of services AWS has to offer.”
That’s from Denneman’s article. That statement tells me they are going to offer a way to use the vast breadth of AWS services with “traditional” workloads that are being migrated to AWS; think ELB, S3, etc.. Also the ability to access them programmatically, not only through the vSphere API, but the AWS API as well. Denneman has more articles on VMware Cloud on AWS, I am hopeful those will have more detail on what else this partnership will mean for customers.
There are two key things that VMware needs to get right for this to work, outside of this not being a duplicate of vCloud Air; pricing and extensibility. No one, and I do mean NO ONE, is going to pay twice the VMware licensing to run on-prem and in the public cloud, they just won’t. I’m glad to see they they are packing everything up and selling it, as opposed to paying for everything separately. Customers hate complex licensing and invoicing. There also needs to be good extensibility to existing and future AWS services that will allow enterprises to start using what AWS has to offer and explore refactoring pieces of their business to take advantage of all the benefits cloud can offer, while still offering on-demand consumption for their “traditional” workloads.