Last week I attended VMworld 2018. Due to my small person living at home, I made a quick trip of it and was in and out of Las Vegas in 13 hours. Even though my stay was short, VMware managed to pack a lot into my visit.
The biggest news was their announcement of the availability of Amazon’s Relational Data Service (RDS) on VMware. RDS is a service that allows customers run and scale their databases in a more automated, easier-to-operate fashion than self-managed databases. This announcement brings the capability of Amazon’s managed databases to customers with on-prem data centers.
This was notable for several reasons.
Amazon and VMware originally announced their strategic partnership in 2016 when they created the ability to run vSphere and VSAN on Amazon Web Services (AWS). This partnership previously brought VMware to Amazon, and now with this RDS announcement they bring an AWS managed service to VMware.
It’s evident that the word “strategic” was not just lip service. As Milin Desai, VMware GM of Cloud Services indicated in the Analyst Day Q&A, VMware’s relationship with Amazon is “unique.” While VMware continues their ‘any app on any cloud’ messaging, the depth of the integrations they have formed with AWS does not appear to be something that will soon be replicated with other IaaS providers.
So what does this strategic partnership mean in light of broader trends? The pervasive wisdom is that the industry’s move to cloud is inevitable over the long-term. This isn’t to say that all workloads will necessarily move, but rather that cloud economics, the environmental impact of shared power and cooling, and the talent and expertise amassed by hyperscale cloud providers all point away from customers competitively running their own infrastructure.
If we accept the premise that cloud is a juggernaut in the future of infrastructure, what do VMware and AWS gain from this partnership?
- If workloads currently reside on-prem but are someday destined to move, then as a cloud provider it makes sense to provide as clean a migration path as possible. In the past AWS provided a place for customers to land a vSphere-based app in the cloud, but now they meeting those workloads where they are. This allows Amazon to begin a) monetizing a broader base and b) establishing sticky services (particularly databases) that will pull customers towards the AWS cloud when they do move.
- From VMware’s perspective as a traditional IT vendor, it makes sense to for them to make hybrid cloud as valuable a proposition as possible for as long as possible; offering customers added convenience is a way to accomplish this. And while the current may be flowing towards cloud, VMware is currently having a moment in the sun. Their customer base is substantial. Their financial results are strong and their stock price is high. They’re often seen as a “safe choice” for enterprise. They are not coming to this partnership from a position of weakness, and they have plenty of assets to leverage.
This brings to mind Michael Coté and Tim Bray’s recent Twitter exchange about cloud being “the longest metaphorical baseball game the IT industry has ever seen.”
Well, compare public cloud revenue to global IT spend and it's not obvious there's even one out in the first.
— Tim Bray (@timbray) September 3, 2018
If this is an eternal game with an incredibly large budget in play on the side on the traditional IT vendor, then the value of the partnership is further illuminated. Traditional software players are very much not dead, and this deepening partnership with the dominant cloud provider is indicative that both VMware and AWS view each other to be substantive partners.
I find the announcement of RDS’ availability on-prem especially interesting when considered in conjunction with Google’s recent announcement of GKE’s availability on-prem at Google Cloud Next in July. While these are inherently different services (database vs. microservice orchestration), they share the similarity of cloud providers bringing their managed services into on-prem environments.
Many moons ago, my high school physics teacher provided some memorable feedback on my somewhat under-developed lab work: “two data points do not a trend line make.” (I had to come in over a lunch hour to re-do the experiment.) I learned my lesson, and as such I’m not proclaiming that these pieces of news constitute a trend of cloud providers moving towards delivering on-prem services, but it is something to watch going forward (especially when viewed in conjunction with hybrid operating models like Azure Stack.)
In all, I came away from VMworld impressed with how the company is navigating a rapidly changing market. (I made one high school teacher happy, and now I’m going to make some others grumpy by ignoring all those English classes and introducing a new idea in my conclusion.) While my main takeaways from this year’s conference were around their cloud-based announcements, it’s worth remembering the context of VMware’s response to other existential threats to their business model. Last year VMware partnered with Pivotal on PKS to respond to shifts away from virtualization as the abstraction layer du jour by creating a managed Kubernetes offering. VMware’s ability to strategically partner to respond to shifting market forces is an impressive competency that should not be underestimated.
Disclosure: VMware comped my VMworld conference ticket. AWS, Google, Microsoft, Pivotal, and VMware are RedMonk clients.