2 ms·
> Containers and HyperVisors/VMs have intrinsically different abilities and different use-cases. Why would you expect one tool to manage both. Can you explain
by acdha 11y ago
> Containers and HyperVisors/VMs have intrinsically different abilities and different use-cases. Why would you expect one tool to manage both.
Can you explain why you believe this to be true? Consider my database server example – if I want to have a MySQL/Postgres service available for our developers to use why would it matter whether it runs inside a VM or a container at the level where I'm saying “Keep 2 copies running on separate hosts for HA” or “Block network access except for hosts in this service group”?
To me, it's a sign of bad system design if the consumers of a service need to know whether it's running on a VM or a container. Only the developers and operators should need to care at that level.
- chris_wot 11y agoTo me, it's a sign of bad system design if the consumers of a service need to know whether it's running on a VM or a container. Only the developers and operators should need to care at that level. This. The advantage of containers is that you isolate and control your resources more precisely - if those using the software running in the container find they have differing requirements, you should just be able to change the resources allocated to the container - in much the same way you would with a normal server.