Hi everyone! First of all, apologies in advance if I sound a bit stupid for asking this but I don't come from a core CS background and don't have much idea about the jargon. So, I am working with a team which is developing courses on Deep Learning and Computer Vision. While exploring the options of autograding, I came across OkPy and nbgrader. Now since we are expecting close to 10k total users and about 200 concurrent users, I decided to have the jupyterhub on the Google Kubernetes Engine. Also, to install opencv and other dependencies, I wrote a dockerfile and as per the Zero to Jupyterhub instructions, I tried using jupyter-repo2docker for generating the docker image and push it on google cloud. The issue is my docker image size is close to 26 GB (I checked it on AWS machine) but my 3 node 1vCPU 100GB boot drive image always ends up showing *no more space left on device*. So these are my doubts: 1. Let me know if I am getting this right or not, but will the docker image be installed separately for each user? Because that will mean close to 300 thousand GB disk space will be required. That's just absurd. Or the docker be installed for the whole hub and the space required for each user will be just for the notebook files? 2. If I understand it right, there are two ways of increasing local disk space on Kubernetes cluster. One is by alloting more nodes and the second is to use SSDs. According to what I have read, the first option is more expensive and is better if we need to increase performance. Which one is a better option and how do I go about setting it up? 3. This one is a really naive query. What does 100 GB of boot disk mean? If I am getting a 100 GB space, that's more than enough to have the docker installed but how do I utilize this? Whenever I do df -h, I get this:
Filesystem Size Used Avail Use% Mounted onoverlay 31G 24G 7.3G 77% /tmpfs 848M 0 848M 0% /devtmpfs 848M 0 848M 0% /sys/fs/cgroup/dev/sdb1 4.8G 82M 4.5G 2% /home/dev/sda1 31G 24G 7.3G 77% /etc/hostsoverlayfs 1.0M 100K 924K 10% /etc/ssh/keystmpfs 848M 708K 847M 1% /run/metricsshm 64M 0 64M 0% /dev/shmoverlayfs 1.0M 100K 924K 10% /etc/ssh/ssh_host_dsa_keytmpfs 848M 0 848M 0% /run/google/devshell Where is the 100 GB going? :( Thanks in advance Vishwesh -- 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.
