James, VP of Engineering & Product Development Leader at nOps, explains a systematic approach to reducing Kubernetes costs.
While the ideal scenario starts with optimizing at the container level followed by infrastructure utilization, James recommends prioritizing based on impact. For most organizations, he suggests focusing on quick wins like moving workloads to Spot instances rather than attempting lower-level architectural changes. The key is to make optimization a continuous process and always start with visibility into your infrastructure.
Watch the full interview: https://ku.bz/xkx3gGmlT
This interview is a reaction to Miguel and Thibault's episode https://ku.bz/_k-Y1jgFS
On the subject of cost optimization and ARM instances, Miguel shared that Adevinta migrated their Kubernetes cluster to support ARM instances and workloads because they are more cost effective. Do you have any practical advice on reducing your Kubernetes costs? I have lots of practical advice, so on reducing kubernetes cost i mean it's really our our subject expertise area over here at nOps you know oftentimes in a greenfield environment we want to start from the bottom up so we want to start optimizing at the container level the lowest level and then we want to start moving our way up making sure that our infrastructure utilization is tuned to get the most out of our infrastructure capacity and then we want to start optimizing for price in actuality though Most organizations don't have the flexibility to start from a greenfield situation. So my advice is to prioritize based on impact. A lot of times, price optimization, meaning that shifting some workloads to spot capacity, using spot best practices, may be a quicker win than trying to re-architect workloads at a lower level. So I say go for the biggest impact with the lowest amount of effort. and make the process continuous. Come back around and figure out what the next biggest priority is and attack that. And always start with visibility. You can't fix if you don't know what the problem is.
