Like many companies, we had to change our database stack midway while scaling our core product Lago, an open-source usage-based billing platform. As we grew more popular, we began ingesting millions of events every minute. And our rudimentary Postgres-only stack wasn’t cutting it. We were suffering heavy load times, impacting our entire app’s performance. After some exploration, we decided to use a distributed ClickHouse instance strictly for our streamed events. Our analytics services were now ...