End of week 1/20

This is the end of the first week on my master’s thesis. I’ve spent most of the week learning Kubernetes and preparing for the master’s thesis in general. Next week I need to plan the rest of the master’s thesis project and start reading some research literature.

Summary of week [1/20]

Here is a short summary of what I’ve done this week:

• Monday
  • Had a start-up meeting at the University for the master’s thesis course
  • Set up my workstation at Elastisys
  • Wrote a blog post
• Tuesday
  • Set up my workstation again, since I received a newer computer :)
  • Started learning Kubernetes through the official tutorial
• Wednesday
  • More Kubernetes learning…
  • Had a meeting with my supervisor on my master’s thesis on how we are going to proceed.
• Thursday
  • More Kubernetes, installed the sock shop
• Friday
  • More Kubernetes learning, investigated methods of collecting metrics in Kubernetes.

I regret that I’ve not found time to read up on some research literature on anomaly detection, so I will have to do that next week. My supervisor sent me some links to read that I need to read next week.

Planning of next week

Next week, I have the following mandatory tasks:

• Send in an updated project description for my master’s thesis to the university
• Send in a detailed time planning for my master’s thesis to the university

I estimate that the above tasks will take approximately 12 hours (1.5 work days) to complete and I need to be done by the end of the week. However, I think it is a good idea to spend 1-2 days reading research literature before updating the project description to get a better grasp on the topic of anomaly detection in container clusters.

Therefore, I estimate I will spend Monday and Tuesday reading research literature, Wednesday and Thursday completing my deliveries to the university, and Friday learning more Kubernetes. It’s probably a good idea to have a meeting with my supervisor on Friday so that we can revise the project planning together before sending it in to the university.

I will probably switch to some Kubernetes-learning during Monday-Thursday as well if I get bored. I haven’t deployed Kubernetes to a cloud provider such as AWS yet, which I need to do.

Something I need to research further during the next week is how to make reproducible environments for testing the anomaly detection methods. I will research this by reading literature and reproducing some work that was performed here at Elastisys during the summer which can be found here and here.