Skip to content

Comment on Show HN: We built a ClickHouse-based logging service

Comments

I started using BigQuery with stream inserts for logging. I have around 300k inserts a day und it costs a whole cent a day for writes and even less for reads (I don’t read much though).

With 3.5 inserts/sec you may equally use SQLite.

What does you ingest setup look like with BigQuery? Are you using something like fluentd to pipe logs over?

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.