Skip to content
This is a repository for complete and easy to use samples that demonstrate the use of Apache Storm on HDInsight
C# PowerShell Java Other
Branch: master
Clone or download
Latest commit 28e0c85 May 12, 2017
Type Name Latest commit message Commit time
Failed to load latest commit information.
HDI3.5 Updating protonJ readme File Apr 6, 2017
lib With unicode support and safer number casting logic (config->long->in… Jan 6, 2017
LICENSE Initial commit Jan 22, 2015 Update Nov 22, 2016
build.bat Integrate with appveyor Aug 19, 2015
build.ps1 Added overall build scripts and nuget restore retry for appveyor Aug 20, 2015
cleanup.bat Added overall build scripts and nuget restore retry for appveyor Aug 20, 2015


This repository contains complete and easy to use examples that demonstrate the power of Apache Storm on Microsoft Azure HDInsight.

The .Net examples are applicable to all HDInsight Windows Storm clusters.

The Java examples are applicable to :

  • 3.3 and below versions of HDInsight Windows
  • 3.4 and below versions of HDInsight Linux Storm clusters.

For version 3.5+ Linux based HDInsight Storm clusters please look under HDI3.5 folder.

Getting Started

Official page - Apache Storm for Microsoft Azure HDInsight

Getting Started page - Getting Started with HDInsight Storm

Repository structure

  1. End-to-end examples - Read more about them below
  2. EventCountExample - Scalability benchmark for Microsoft Azure EventHubs/Apache Kafka on Microsoft Azure HDInsight
  3. IotExample - Internet of things - connected cars scenario
  4. RealTimeETLExample - Web request aggregation into HBase
  5. SCPNetExamples - Examples that show API usage of SCP.Net
  6. lib - Java dependencies
  7. scripts - Automation scripts
  8. Templates - Azure services templates
  9. tools - Helper tools

Running the examples

Each of the examples folder contains a run.bat that does the following:

  1. Prepare (using prepare.ps1)
  2. Creation of required services (and resources) in Azure
  3. Updating project properties and templates
  4. Build (using build.ps1)
  5. Building the source code and use the properties created in previous step
  6. Execute (using execute.ps1)
  7. Upload jars and create topology packages (as needed)
  8. Submitting the topologies to the HDInsight Storm cluster

An existing run configuration is first detected to provide idem-potency in creation of Azure resources. This allows you to re-run the run.bat in case of some unexpected failure.

Building the examples

If you just want to build the source code you can use the build.bat in each example folder to build that example.

Re-running the example

If you modified some source and wish to re-deploy the topologies, you can build using build.bat and submit the topologies using execute.ps1


There are multiple end-to-end examples included in this repository that showcase different topologies that you can build and run against HDInsight Storm. HDInsight provides users the option to write applications in language of their choice and exposes Storm REST APIs as well as additional web interfaces for remote topology submission and management.

Internet of Things Example (Java)

An internet of things example that demonstrates connected cars scenarios. The example showcases the reading of vehicle events from Event Hub, referencing Azure DocumentDB for vehicular data and then finally storing the computations in Azure Storage using the WASB driver on HDInsight clusters.

Read more about this example here: IoT example using Azure EventHubs, Storm, Azure DocumentDB, Azure Storage/WASB

EventHubs/Kafka Scalability Example (Java & SCP.Net)

A scalability benchmark that demonstrates throughput from EventHub/Kafka and stores the event counts in SQL Azure.

Read more about this example here: Event Count example using Azure EventHubs, Storm, SQL Azure

Real Time ETL Example (SCP.Net)

A real time ETL example with EventHubs, Storm and HBase that has two topologies.

  • The first topology Event Sender Topology generates Web Request Logs and pushes them to EventHubs.
  • The second topology Event Aggregation Topology reads from EventHubs, calculates aggregations("count() group by") and then stores the computations in a HDInsight HBase cluster.

Read more about this example here: Real Time ETL example using Azure EventHubs, Storm, HBase & SCP.Net


For writing Storm applications in C# using SCP.Net refer

SCP.Net Examples

Check out more SCP.Net Examples

This directory now contains many SCP.Net examples that show the APIs available in SCP.Net NuGet package. Follow the instructions in the above directory to deploy these examples to your HDInsight Storm cluster.

The HybridTopologyHostMode in SCPNetExamples/HybridTopologyHostMode has examples on all the different types of hybrid modes.

Topology, Spout Bolt templates

Templates to connect to various Azure services:

  • EventHubs
  • SQL Azure
  • DocumentDB
  • HBase
  • Azure Websites (SignalR)

Templates for Azure services

Azure Services and HDInsisght Helper PowerShell Scripts


Build status

Build status

You can’t perform that action at this time.