White Papers

White Paper: How Kubernetes Saved OpenStack

Issue link: https://read.uberflip.com/i/1387901

Contents of this Issue


Page 0 of 1

How Kubernetes Saved OpenStack © 2020 Mirantis Inc. All Rights Reserved. Information is subject to change. | www.mirantis.com When virtualization was big and cloud was born, the idea of open source private cloud seemed powerful. VMware had shown that organizations wanted more efficient ways to utilize hardware. Both VMware and AWS enabled self-service and rapid provisioning of virtual machines. But both could be uncomfortably expensive at scale, so the thought was to create something that could be used on-prem to provide a self-service environment for users who wanted to create their own VMs. A properly running OpenStack cloud could provide resources solidly enough for entire telecommunications systems to run on it; in fact, an entire branch of networking, Network Functions Virtualization (NFV), evolved to replace single-purpose hardware such as firewalls and load balancers with virtual machines, managed by OpenStack more often than not. But while the idea of open source private cloud remains powerful, there came a point where containers pointed the way to various other kinds of efficiency, and Kubernetes emerged and drew focus away f rom virtualization. Containers had a lot of advantages over virtual machines, after all, such as better portability and smaller sizes. Somewhere in the back of many developers' heads there grew a notion that everything should be containerized, whether or not it made sense for the particular application. People began to talk about containerizing legacy applications, and even considered turning Virtual Network Functions (VNFs) (the workloads of NFV) into Containerized Network Functions (CNFs). Developers were twisting themselves (and their code) into knots just to containerize their workloads Part of the reason containers and Kubernetes gained such traction is that OpenStack, as powerful as it is, was in a sense a weaker vessel, in that it was hard to deploy, operate, and maintain. Standing up an OpenStack cloud be a multi-week (or month) job for multiple staff members. And upgrades could be so f raught with danger (or at least perceived danger) that production installations often didn't upgrade at all, staying 2, 6, even 10 versions behind rather than take the chance to get the newer functions available in OpenStack.

Articles in this issue

Links on this page

view archives of White Papers - White Paper: How Kubernetes Saved OpenStack