Operations | Monitoring | ITSM | DevOps | Cloud

During the AWS summit in London, Renee Hunt, the CTO of Compare the Market, shared their journey of migrating from EC2 to EKS and the obstacles they faced along the way. As I listened to their story, I couldn't help but think about how Qovery could have greatly streamlined its migration process; here is my take on the subject.

One of the first considerations for FinOps teams trying to lower their public cloud spend is investing in long-term savings vehicles available from their Cloud Service Provider. These programs can provide customers with upwards of 72% savings off on-demand prices, in return for a 1-to-3-year usage commitment, so it’s pretty common that we see them in use by our customers.

In part I of this blog series, we understood that monitoring a Kubernetes cluster is a challenge that we can overcome if we use the right tools. We also understood that the default Kubernetes dashboard allows us to monitor the different resources running inside our cluster, but it is very basic. We suggested some tools and platforms like cAdvisor, Kube-state-metrics, Prometheus, Grafana, Kubewatch, Jaeger, and MetricFire.

Kubernetes (k8s) adoption has skyrocketed since its 2014 introduction, becoming one of the most popular open source container orchestration platforms for its power and flexibility. K8s reduce costs by improving efficiency, optimizing resource use, and eliminating redundancies. But cost savings in Kubernetes can be tricky to maintain. In fact, a 2021 survey of 178 organizations showed that costs associated with k8s can actually increase from insufficient monitoring, resulting in overspend.

Cloud infrastructure and the applications it hosts generate millions of rows of raw data in short periods. Viewed in the context of everyday business activities, most of the data makes little sense. So many companies amass mountains of this type of data in data lakes and data warehouses, hoping to figure out how to use it later but end up never tapping into it.