Netflix's distributed Data Pipeline
Java GAP

Suro: Netflix's Data Pipeline

Suro is a data pipeline service for collecting, aggregating, and dispatching large volume of application events including log data. It has the following features:

  • It is distributed and can be horizontally scaled.
  • It supports streaming data flow, large number of connections, and high throughput.
  • It allows dynamically dispatching events to different locations with flexible dispatching rules.
  • It has a simple and flexible architecture to allow users to add additional data destinations.
  • It fits well into NetflixOSS ecosystem
  • It is a best-effort data pipeline with support of flexible retries and store-and-forward to minimize message loss

Learn more about Suro on the Suro Wiki and the Netflix TechBlog post where Suro was introduced.

Master Build Status

Pull Request Build Status


NetflixGraph is built via Gradle ( To build from the command line:

./gradlew build

See the build.gradle file for other gradle targets, like distTar, distZip, installApp, and runServer.

Running the server

You can run the server locally by just running ./gradlew runServer.

More more advanced usage you may wish to run ./gradlew installApp and then:

cd suro-server
java -cp "build/install/suro-server/lib/*" -m conf/routingmap.json -s conf/sink.json -i conf/input.json

To enable basic logging you can downloaded slf4j-simple-1.7.7.jar and copy it into suro-server then run:

cd suro-server
java -cp "build/install/suro-server/lib/*:slf4j-simple-1.7.7.jar" -m conf/routingmap.json -s conf/sink.json -i conf/input.json


We will use the Google Group, Suro Users, to discuss issues:!forum/suro-users