Experiments with connection poolers on Kubernetes for Postgres Operator

This year, my teammates and I solved a CPU throttling issue that affects nearly every container orchestrator with hard limits, including Kubernetes, Docker, and Mesos. In doing so, we lowered worst-case response latency in one of Indeed’s applications from over two seconds to 30 milliseconds. In this two-part series, I’ll explain our journey to find the root cause and how we ultimately arrived at the solution.

KubeDB by AppsCode is a production-grade cloud-native database management solution for Kubernetes. KubeDB simplifies and automates routine database tasks such as provisioning, patching, backup, recovery, failure detection, and repair for various popular databases on private and public clouds. It frees you to focus on your applications so you can give them the fast performance, high availability, security and compatibility they need.

- Define a path for error handling
- Have a plan for recovery
- Always set timeouts on remote system calls
- Retry on timeout
- Use circuit breaker
- Don't handle timeouts like a failure
- Don't invoke remote systems inside transactions
- Use smart batching
- All APIs MUST be idempotent
- Define response time and throughput SLAs explicitly and code to adhere to them
- Define and limit batch APIs
- Think about Observability up-front
- Cache aggressively
- Consider unit of failure
- Isolate external domain objects at the edge of the system
- Sanitize input at every edge
- Never commit credentials

A lightweight and ultra-fast tool for building observability pipelines

So just for fun, let's see what the absolute bare minimum "Kubernetes cluster" actually looks like.