Backend

Eng teams defaulting to DDOG was peak ZIRP. I’m glad I steered our team clear of this budget drainer.

enzo's dad 🇬🇷 in LA 🌴 eng @ fanatics.live ⚾️ alexloukissas.com

wins

biotech software engineer // farcaster OG // fid 589

Prometheus /w Grafana. You can run it yourself for very little cost. Can even do almost infinite retention with S3 glacier. For otel traces, we have been really happy with honeycomb (also predictable, low cost).

Great resources here https://github.com/prometheus-community/helm-charts/tree/main/charts/kube-prometheus-stack, with a ton of ready to use dashboards. Prometheus operator also very useful.

Yes, you will need to DIY some stuff. But your CFO will high-five you so hard, your hand will hurt.

Finally: here is how you can plug in various persistence backends like S3: https://thanos.io/ (epic name too!)

Definitely +1 on this for me, Prometheus / grafana have worked great for us where I'm working. The maintainers even create grafana cloud (paid managed version), and migrate to your own deployment if it gets big.