Hi folks, relative newcomer here. :) I'm working on using Kubernetes to create a render farm for my company. One of the things I need to be able to do is measure machine utilization so that I can bill our customers. I looked into the heapster docs, but that doesn't seem to be the correct tool - it seems to be focused on real-time monitoring, whereas what I need to know is the total resource usage for a job that may last 10 hours or more. (More accurately, I want to know how much AWS is going to charge us, but I need to know it on a per-container basis, since multiple containers for multiple customers may be running on a single node.)
The approach I've been looking at is to get the docker stats by talking to the node directly. That is, given a job, query to find the pod metadata, then talk to the pod to get the docker stats for the container. I haven't quite gotten this to work yet. Is there a better approach? -- You received this message because you are subscribed to the Google Groups "Kubernetes user discussion and Q&A" group. To unsubscribe from this group and stop receiving emails from it, send an email to kubernetes-users+unsubscr...@googlegroups.com. To post to this group, send email to kubernetes-users@googlegroups.com. Visit this group at https://groups.google.com/group/kubernetes-users. For more options, visit https://groups.google.com/d/optout.
