The goal is to provide a cost-effective way to monitor postgres and data pipelines at scale. Onboarding a new db takes minutes. Controls entire purpose is to answer the question: "Are customers being affected?"
I have experience operating pretty large data clusters and pipelines (100's of thousands of instances). Control takes these learnings and exposes them in the product:
- SLO based alerts by default model that errors exist in reality - Verdicts provide health classifications, instead of requiring the operator to read through 10's or hundreds of metrics. - Flat pricing based on instances. - The control observability protocol is pushed based and no customer or sensitive data is stored in control.
How is everyone monitoring their databases and pipelines today? If you're using a big vendor, what's working for you? What's not?
Looking forward to everyone's feedback,
Thank you!
Danny