Recently, I gave a presentation at MongoDB Silicon Valley User Group titled “Fluentd loves MongoDB”.
For those who don’t know what Fluentd is…Fluentd is a log collector daemon written in Ruby. It receives logs as a JSON stream, buffers them, and sends them to other systems like MongoDB.
We originally developed it to solve the problem of data collection for our own service. Since open-sourcing it last October, it’s been deployed at many organizations including SlideShare/LinkedIn, ContextLogic, Viki, NHN, and many others.
Fluentd and MongoDB work well together because both handle data as JSON objects. A number of Fluentd’s users deploy MongoDB for real-time analysis whilie delegating batch analysis to their Hadoop backend (or Treasure Data).