How to select Kubernetes platform for deployment at scale

Selecting an appropriate Kubernetes platform for your organization is a challenging task. This talk covers some important elements that you should consider from a Day 0, Day 1 and Day 2 perspective. We will also go over cost optimization, developer productivity and delivering business value

Topics: Day 0 - Its design time: Analyzing existing environments:

Networking
  Software Defined Networking
  Container Networking Interface
  CNI plugins

Storage
  Software Defined Storage
  Container Storage Interface
  CSI plugins

Compute
  Bare metal vs virtualization
      What hypervisors do you already use?
  HA for master nodes
  HA for worker nodes

Day-1, Developing playbooks for cluster install process Monitoring System / Host level metrics Monitoring container level metrics Monitoring applications

Day-2 Operations: Keeping lights on Patching system for security Version upgrades Backups HA-DR: what is the shared responsibility model? How to recover from a disaster?

Developer productivity: Who writes the yaml files? How much yaml should they learn? Who gets to define new LoadBalancers, Ingress rules etc Securing access to Kubernetes dashboard

Helm charts: single command deploy 
Istio: implementing service mesh architectures.
	Injecting failures for Chaos engineering purpose
	Circuit-breakers you don’t have to write.
	Canary your changes so you don’t have to get paged Skillset evaluation    Kubernetes is hard: steep learning curve if you choose to operate it.