Will Kubernetes finally make OpenStack easier to use? Will serverless computing render the OpenStack and Kubernetes unnecessary? Will the Open Container Initiative ever finish the 1.0 version of its container runtime?
These were all questions that were grappled with in the latest TNS Pancake Podcast, held in Mountain View California’s Computer History Museum during OpenStack Silicon Valley 2016. Panel guests included Google product manager Craig McLuckie, CoreOS vice president of engineering Mike Saparov, Mirantis co-founder and CEO Alex Freedland, and Iron.io. co-founder and CEO Chad Arimura. TNS founder Alex Williams hosted the panel.
Freedland launched the discussion with a mention of how containers have come to take a place within the OpenStack community. Initially, OpenStack developers may have viewed containers as the minivans of the software development world, built for transporting workloads from one place to another. “You understand that no matter what you are delivering, whether it’s OpenStack or Kubernetes, it can be done in a similar way. That I think is the massive change that is happening as a result of the Docker and Kubernetes revolution. OpenStack majorly benefits from that,” Freedland said.
Serverless computing also had its place in the discussion surrounding orchestration, which Arimura explained has helped to usher in the PaaS of the future, “The concept of serverless is not that there aren’t any servers, it’s that you’re not thinking about them when you’re developing software. I think from that standpoint, you really are thinking about PaaS 2.0 or the next generation of PaaS.”
McLuckie highlighted the importance of growing the Kubernetes ecosystem, hoping that any emerging organizations looking to focus on Kubernetes will engage users where they are at in the process of implementing container orchestration platforms. Saparov went on to highlight the issues existing organizations focusing on Kubernetes are facing, “They have technology challenges, headcount and talent gap challenges, and process challenges. Whether it’s Kubernetes, cloud storage, public cloud, or OpenStack. You have to have a way to engage with them to help address all of these parameters. It’s really a people, processes, and technology combination for all three.”
McLuckie wrapped up the discussion by highlighting three of the key reasons Google brought Kubernetes to the Cloud Native Computing Foundation. After fairness and maintaining a strong sense of technical identity, McLuckie implored that, “Most importantly, and what I want to reach out to to everyone here, is that we wanted to create a foundation that had legitimate accountability to the end users, that wouldn’t be driven by the vendors.”
Feature image: (Left to right) Craig McLuckie, Mike Saparov, Alex Freedland, Chad Arimura, Alex Williams.