I played a bit with Azure Event Hubs and Stream Analytics before Azure IoT Hub came out and the integration was fairly straight forward. My goal was to get the data into Azure for long-term storage/archival and the ability to feed the data into Azure Machine Learning and other tools.
The integration was fairly straightforward: I took the webhooks SmartApp by @harper and adjusted it to post the data to my Event Hubs endpoint which is automatically made available by Azure. Once the data hit Event Hubs, I used Stream Analytics to transfer the data into Azure Table storage and then built a simple web Azure Web App to visualize and trend the data.
After doing all that, I was questioning why I was using Azure Event Hubs for this since it seems really overkill for a small proof of concept. Keep in mind Event Hubs can log millions of events per second and I was doing this for just my house.
Since Event Hubs are a bit expensive for what I was getting out them, I was curious if there were other tools which might work for my particular application. I ended up changing over to Azure Queues which still support 2000 events per second and are super cheap (basically the dirt cheap storage costs). There are a lot of tradeoffs going this route which probably don’t make sense for many event driven applications, but it worked fine for my application which was mostly archival purposes.