Now that we know how to develop a Storm topology, let's go over how to package it up into a JAR file and deploy it onto a cluster.
In a terminal, navigate to the trucking-iot-demo-storm-on-scala directory and run:
After installing some dependencies, this script runs the command
sbt assembly under the hood in order to produce an uber jar, housing your topology and all of the dependencies. The jar is saved to
If you ran the command in the previous section, then the topology was build and then also deployed.
Feel free to open that script to see what it is doing. Notice that under the hood, it runs a command that looks like the following.
storm jar trucking-iot-demo-storm-on-scala/target/scala-2.12/trucking-iot-demo-storm-on-scala-assembly-1.1.0.jar com.orendainx.trucking.storm.topologies.KafkaToKafka
storm will submit the jar to the cluster. After uploading the jar,
storm calls the main function of the class we specified (com.orendainx.trucking.storm.topologies.KafkaToKafka), which deploys the topology by way of the
Congratulations! You now know about the role that Storm plays in a real-time data pipeline and how to create and deploy a topology from scratch.