System requirements

These are the versions of dependencies that the sample applications were tested against.

  • Oracle JDK 1.7 (64 bit version).
  • Apache Storm 0.9.
  • Apache Zookeeper (required by Storm).
  • Apache Pig 0.12.
  • Redis 2.8.4.

Build environment

  • NetBeans 7.4.
  • Apache Maven – version that is included with Netbeans 7.4.

Steps required to run sample

Dependency configuration

  • Ensure that dependencies are properly configured and running.
    • Apache Zookeeper needs to be running
      • Navigate to zookeeper$version/bin.
      • Start ZooKeeper with ./zkServer.sh start.
      • On Windows use zkServer.cmd start.
    • Storm needs to be running.
      • Navigate to storm-$version/bin.
      • Run the following commands to start Storm.
        • ./storm nimbus &
        • ./storm supervisor &
      • To configure Storm on Windows please refer this blog post.
      • On Windows use the short version of paths that contain spaces (for instance, c:\program files\... should be shortened) when configuring JAVA_HOME. We had trouble with paths that contain spaces when running Storm.
    • Redis needs to be running.
      • Quick start information for Redis is available here.
    • Installing Pig is optional. Setup details are here. Installing Hadoop is not required for our purposes since we will just run Pig in local mode.
  • For additional details on getting started with Zookeeper, Storm, Pig and Redis, please refer to getting started information on their project pages.

Optional steps

  • Building and running generatedata is optional since pre-generated data is available in the batch folder.
  • Running MapReduce using calculate-max.pig available in the batch folder is also optional. A pre-generated view for the batch layer is also made available under the batch folder. The file containing this view is named part-r-00000.

If you choose to run the Pig script you can run in local mode as below.

pig -x local calculate-max.pig

Required steps

  • Build the utils project. The realtime and query projects depend on utils. utils can be built with Maven or an IDE such as Netbeans or Eclipse.
  • The Storm project available under the realtime folder. realtime can be built with Maven or an IDE such as Netbeans or Eclipse. Once built you can deploy the resulting jar file on your local Storm development cluster using the command below. Please note that minor code changes may be needed to deploy to an actual non-development Storm cluster. We tested with a local development cluster.

storm jar realtime-1.0-SNAPSHOT-jar-with-dependencies.jar com.syncfusion.realtime.MaxTrackerTop

  • Once the storm job is completed (it shuts down automatically), or even when it is running you can build and run the query sample. The query sample writes top ten pressure values form the realtime and batch views to the console. query can be built with Maven or an IDE such as Netbeans or Eclipse.