|
The company I work for currently has about ~ 150 CentOS 7 VMs running on Nutanix AHV. Running all sorts of things from simple Apache and Nginx deployments, to things like Jira and Confluence, and Nagios monitoring with a Master + Slave setup, Gitlab, and a few other bits and bobs. I've been playing with Podman a bit lately and getting a feel for it, but I don't really have a strong knowledge of containers yet besides deploying them, destroying them, and basically using them like VMs and adding persistent storage and "stuff like that" (I'm really a novice to the whole Containers/Kubernetes/whatever thing). But... CentOS 7 is getting perilously close to EOL and my seniors are talking about migration candidates such as Rocky, Alma, RHEL, Debian, Ubuntu, SUSE - They haven't decided on anything concrete yet. But assuming its Alma for example... How complex would it be to run these types of workloads as containers, and how difficult would be the process of going from VM > Container from a Nutanix cluster running AHV to say... A Nutanix host running lots of containers and a few VMs? Is it a good move? Does it make more sense from an infrastructure perspective? From what I understand, instead of assigning more RAM or CPU like we do to a VM when its required for XYZ reason (or deemed required), we could instead scale up the number of containers to achieve similar results. Is that right? Sorry if this isn't clear. Just looking for guidance and trying to get a better idea. Thanks! |