New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Both are distinct technologies, both for separate use cases. One could often accomplish a goal by using either of them, but do know that process-containers are not made for stateful workloads.
So I would recommend choosing your tech stack based on your use case rather then trying to get a highscore on buzzword-bingo. The reason most people in here talk about VPS is because they need something that's easy to manage without a massive convoluted bloated overhead for no reason. There is no reason to install wordpress in k8s, and it's totally moronic to setup postgres or mysql inside k8s for example.
Below is a diagram with a quick overview of all the things that fall under the container buzzword umbrella.
Source:
https://www.youtube.com/watch?v=0z3yusiCOCk
KVM with docker is my preferred route. KVM because I like picking my own images etc. Docker because you can deploy a grouping of services via text file basically (docker-compose). And yeah that's a overhead as Virtual says...but the clean simplicity of creating & destroying docker stuff appeals
Haven't personally found a use for k8s...seems more geared towards loadbalancing and orchestration. I'll cross that bridge when needed
You’re conflating terms a bit, but I get the gist of what you’re asking. And +1 to @havocx. Is there overhead to running basic stuff (like a webserver / backup stack) in a container? Sure.
But the convenience of being able to simply transfer a single directory to a new device and spin up your stack in a matter of minutes is unmatched.
@ouvoun
If you are using native networking, there should be ~0% difference between 'bare metal' and 'docker'.
Networking, sure, if you have it set to use host. But IO is definitely impacted by containerization, and memory usage could be a concern in resource constrained scenarios.
And how often do you do that?
My k8s nodes spin up new pods to scale every few minutes.