[kevmo314]

The wildest part is they’ll take those massive machines, shard them into tiny Kubernetes pods, and then engineer something that “scales horizontally” with the number of pods.

[jesse__]

Yeah man, you're running on a multitasking OS. Just let the scheduler do the thing.

[dgxyz]

Yeah this. As I explain many times to people, processes are the only virtualisation you need if you aren’t running a fucked up pile of shit.

The problem we have is fucked up piles of shit not that we don’t have kubernetes and don’t have containers.

[mystraline]

Its all fun and games, until the control plane gets killed by the OOMkiller.

Naturally, that detaches all your containers. And theres no seamless reattach for control plane restart.

[zacmps]

Until you need to schedule GPUs or other heterogenous compute...

[pnt12]

This is especially aggravating when the os inside the container and the language runtimes are much heavier than the process itself.

I've seen arguments for nano services (I wouldn't even call them micros services), that completely ignored that part. Split a small service in n tiny services, such that you have 10(os, runtime, 0.5) rather than 2(os, runtime, x).

[SpaceNugget]

There is no os inside the container. That's a big part of the reason containerization is so popular as a replacement for heavier alternatives like full virtualization. I get that it's a bit confusing with base image names like "ubuntu" and "fedora", but that doesn't mean that there is a nested copy of ubuntu/fedora running for every container.

[andai]

I had to re-read this a few times. I am sad now.

[cyberpunk]

To be fair each of those pods can have dedicated, separate external storage volumes which may actually help and it’s def easier than maintaining 200 iscsi or more whatever targets yourself

[ahartmetz]

I think my brain hurts

[jayd16]

I mean, a large part of the point is that you can run on separate physical machines, too.