Operate first Jupyterhub application provides a Jupyter notebook environment for python development and running machine learning workloads. In this repository, we analyze the data generated from the application hosted on the Operate First cluster on Openshift.
For setting the resource tiers for the JH application, we also need to identify the fixed amount required to start up the user pods. These fixed costs are the minimum required to start a pod, load image libraries, and then start a notebook kernel.
We should use this issue to discuss if there are other fixed costs and to find the best way of identifying them.
One way is to look at the consumption of an idle pod, but how do we define idle?
The other way could be to see the consumption of the pod just after it starts and stabilizes, but how do we know the libraries and the kernel are loaded at that point?
For setting the resource tiers for the JH application, we also need to identify the fixed amount required to start up the user pods. These fixed costs are the minimum required to start a pod, load image libraries, and then start a notebook kernel.
We should use this issue to discuss if there are other fixed costs and to find the best way of identifying them.
cc @tumido