We fight with the definition around the term "serverless" (which I talked about in the past here) and dealing with the continuous hype cycle that brings these things to the forefront of the news. We see the talk about containers and PaaS. The onslaught of magic quadrants and surveys would have you believe that you are living in the past if you aren’t already running all public cloud and serverless. Don’t even get me started on unikernels and that vein of the market.
The reason to start there is that we have to look at the points where the hype cycle lands in reality. When we talked about the rise of distributed computing, the dawn of Windows NT became a turning point.
NT to Nano: Look How Far We’ve Come
Azure Stack is in Technical Preview 3 at the time of this writing. Production availability will be in a matter of months (estimated Q2 2017) and while we look down the layers to the IaaS, Windows Nano is becoming a pretty popular way to move up the stack. As much as the infrastructure matters, so much more attention needs to be put up in the application and workload tiers. This is where Microsoft is pivoting nicely as well.
Containers are now being delivered on Windows. Windows Nano server is becoming a hot commodity as we look to drive up the adoption of immutable and composable infrastructure. As the Azure Stack lands in 2017, it is inevitable that the market will be well warmed up to the data center operating system concepts. To quote the investment term, I’m long on Microsoft.
Are We Ready for the Next Steps? Yes!
The cultural shift towards embracing DevOps methodologies is underway. Not in the way that we are all becoming like Netflix or LinkedIn, but in a way that we have seen that a shift needs to occur. The 80s and 90s did show some potential for this idea of Rapid Application Development and the iterative application development methodologies that were coming out of some of the university campuses and startups.
I’ve been speaking at events and working with the community on enabling the embrace of this next generation where thin is in and we can rethink how we achieve the goals. Even if the goals have not shifted, the way in which we can satisfy the business requirements has. We are clearly ready for this. We just need to let it evolve.
Big Containers, Small Operating Systems, and Velocity
The coming months and years are going to be dominated by the investigative stages. This is where we make the same mistakes that we did with early stages of virtualization. This is the natural evolution as the next wave comes in. Along with all of these oversized containers and more agile, tiny operating system environments, we are increasing the velocity of IT.
The velocity increase comes because of the composable nature of the next generation of IT. Whether it’s Docker, Rocket, CoreOS and Kubernetes, DC/OS, Photon, or any combination of them, the velocity will also be met with the need for autonomic systems in these environments. The reason that we see Google able to scale at incredible rates is because they have designed autonomic systems to keep them performing at scale.
You can see the pattern developing. Before we bring it all together, it is also important to take one last stop in the journey towards autonomic computing to enjoy a little game in our next post.