InfluxDB vs. RethinkDB vs. Riak

Server and application metrics are all stored in InfluxDB for capacity planning and anomaly detection.

All our metrics go through InfluxDB, both applicative and system

To track time-series of course, utilizing few retention rules and continuous queries to keep time-series data fast and maintanable

InfluxDB ingests information from various sources (mostly Telegraf instances) into one place for monitoring purposes.

We use InfluxDB as a store for our data that gets fed into Grafana. It's ideal for this as it's a lightweight storage engine that can be modified on the fly by scripts without having to log into the server itself and manage tables. The HTTP API also makes it ideal for integrating with frontend services.

Main database, using it in multiple datacenters in an active-active configuration.

Sharded and replicated storage, NoSQL with joins

As a boring document oriented database with safe defaults.

High-speed update-aware storage used in our region server infrastructure; provides a good middle layer for storage of rapidly modified information.