Storing IOT Data in Azure: SQL vs Cosmos vs Other Methods

ChandanJha 

There are a number of choices that you can make in order to make this architecture more scalable and resilient. Without fully understanding the consumption use-case, it's generally thought that IoT data does not lend itself very well to relational data stores as the raw storage layer, and that is why you see a lot of the reference architectures with a data lake, or a data lakehouse (Data Lake combined with ACID capabilities, such as Databricks Delta) as the central repository, and then other data stores being added based on usage patterns (Azure Synapse analytics for aggregate reporting, for example)..

There are some very good examples of reference architectures for IoT analysis available here: Azure IoT Reference Architecture (microsoft.com) 

Depending on what you want to do with the collected data, you can also look to incorporate tools such as Azure Time Series Insights (Gen2 use cases - Azure Time Series Insights Gen2 | Microsoft Docs) or even Azure Metrics Advisor (Azure Metrics Advisor – Root-Cause Analysis | Microsoft Azure) to help add value to the overall solution.

I hope this helps.

Thanks.

Ted Malone

Microsoft - West Enterprise Commercial