Amazon has made the AWS IoT Analytics service generally available, which provides advanced analysis of data collected from IoT devices. At the AWS re:Invent conference last year, Amazon released a first preview version of AWS IoT Analytics, and this service is a part of a broader IoT-focused push by Amazon since this conference.
During the time of the preview release and now, Amazon has improved the service. Randall Hunt, senior technical evangelist at AWS, stated in the blog on the release of the AWS IoT Analytics service:
Iterating on customer feedback received during the service preview, the AWS IoT Analytics team has added a number of new features including the ability to ingest data from external sources using the BatchPutMessage API, the ability to set a data retention policy on stored data, the ability to reprocess existing data, preview pipeline results, and preview messages from channels with the SampleChannelData API.
AWS customers can leverage AWS IoT Analytics to ingest data using "channels". The channels serve as an entry point for external sources or existing IoT Core MQTT topics that send messages to them. Next, the pipelines allow processing of data through various steps, so-called "activities", to perform complex transformations ("wrangling") on the message content by invoking Lambda functions, or enrich with data from IoT Core. The output of wrangled data goes to a data store, optimized for data queries. Customers can create data sets from querying the data stores manually or using a recurring schedule. Finally, with "notebooks", which are Amazon SageMaker hosted interactive Jupyter notebooks, customers can analyze their data with custom code and even build or train ML models on the data.
Source: https://aws.amazon.com/iot-analytics/
Users can set up an AWS IoT Analytics service instance through the console or the Command Line Interface (CLI). Subsequently, they will go through a process supported by a wizard to specify the channels, create a pipeline with activities, and data store to output the results of the pipeline. Furthermore, the wizard will guide the user to create a dataset, and a model from the data into a SageMaker powered Jupyter notebook. Users can benefit from templates AWS provides such as anomaly detection or output forecasting. Moreover, they can build visualizations using QuickSight with their datasets.
Source: https://aws.amazon.com/blogs/aws/iot-analytics-now-generally-available/
The release of AWS IoT Analytics as part of Amazon's IoT-focused push will compete with several established vendors like Microsoft, Google, and IBM, who have made quite significant investment in their IoT offerings. Microsoft provides the Azure IoT Suite, which consists of Azure services for connectivity, like the IoT Hub and analytics such as Stream Analytics and Machine Learning. Furthermore, Google with their recent release of the Cloud IoT Core delivers a complete IoT solution. The IoT core offers not only a way of managing millions of connections to devices, but also integration with the Google Could Platform data analytics services like Google Cloud Dataflow, Google Cloud Machine Learning Engine, and BigQuery. Finally, IBM has Watson Internet of Things (IoT), which too offers capabilities for connectivity and analytics.
Amazon prices the AWS IoT Service as pay-as-you-go, and there is a free tier for their customers that is valid for the first 12 months. Furthermore, the service is now available in US East (N. Virginia), US West (Oregon), US East (Ohio), and EU (Ireland), with availability in other regions planned for a future release. More details about AWS IoT analytics are available in the documentation.