Amazon IoT Analytics automates all the steps required to run and operationalize analytics on IoT data. Amazon IoT Analytics automates the difficult steps that are required to analyze data from IoT devices. Amazon IoT Analytics can accept data from any source including Amazon Kinesis, Amazon S3, or third party tools, using a BatchPutMessage API and is fully integrated with Amazon IoT Core so it is easy to collect data and begin performing analytics. First, you define a channel by using MQTT topic filters to specify only the data you want to store and analyze. Once the channel is set up, you configure a pipeline to process your data. The pipeline can perform data transformations, execute conditional statements, and enrich messages with data from external sources.
After processing the data, Amazon IoT Analytics stores it in a time-series data store for analysis. Then, you can run ad hoc or scheduled queries using the built-in SQL query engine to answer specific business questions, or perform more sophisticated analysis and machine learning.
Ingest data from any source including Amazon IoT Core - Ingest data directly from Amazon IoT Core to Amazon IoT Analytics. Or, use the BatchPutMessage API to send your data to Amazon IoT Analytics from Amazon S3, Amazon Kinesis or any other source. With Amazon IoT Analytics' full integration with Amazon IoT Core and the API, it is easy to receive messages from connected devices as they stream in.
Collect only the data you want to store and analyze – You use the Amazon IoT Analytics console to configure Amazon IoT Analytics to receive messages from devices through MQTT topic filters in various formats and frequencies. Amazon IoT Analytics validates that the data is within specific parameters you define and creates channels. Then the service routes the channels to appropriate pipelines for message processing, transformation, and enrichment.
Cleanse and filter – Amazon IoT Analytics let you define Amazon Lambda functions that can be triggers on when Amazon IoT Analytics detects missing data, so you can run code to estimate and fill gaps. You can also define max/min filters and percentile thresholds to remove outliers in your data.
Time-series data store - Amazon IoT Analytics stores the device data in an IoT optimized time-series data store for analysis. You can manage access permissions, implement data retention policies and export your data to external access points.
Store processed and raw data - Amazon IoT Analytics stores the processed data and also automatically stores the raw ingested data so you can process it at a later time.
Run ad hoc or scheduled SQL queries - Amazon IoT Analytics provides a built-in SQL query engine so you can run ad hoc or scheduled queries and get results quickly. For example, you may want to run a quick query to find out how many monthly active users there are for each device in your fleet.
Time-series analysis - Amazon IoT Analytics supports time-series analysis so you can analyze the performance of devices over time and understand how and where they are being used, continuously monitor device data to predict maintenance issues, and monitor sensors to predict and react to environmental conditions.
Hosted notebooks for sophisticated analytics and machine learning - Amazon IoT Analytics includes support for hosted Jupyter Notebooks for statistical analysis and machine learning. The service includes a set of pre-built notebook templates that contain Amazon Web Services-authored machine learning models and visualizations to help you get started with IoT use cases related to device failure profiling, forecasting events such as low usage that might signal the customer will abandon the product, or segmenting devices by customer usage levels (for example heavy users, weekend users) or device health.
You can do statistical classification through a method called logistic regression. You can also use Long-Short-Term Memory (LSTM) which is a powerful neural network technique for predicting the output or state of a process that varies over time. The pre-built notebook templates also support the K-means clustering algorithm for device segmentation, which clusters your devices into cohorts of like devices. These templates are typically used to profile device health and device state such as HVAC units in a chocolate factory or wear and tear of blades on a wind turbine.
Bring your custom container - Amazon IoT Analytics will import your custom authored code containers, built in Amazon IoT Analytics or a third party, such as Matlab, giving you more time to focus on what sets you apart from your competition. No need to recreate your existing analyses created in third party tools. Simply import your analyses container on Amazon IoT Analytics and execute it as needed.
If you are using Jupyter Notebooks, simply create an executable container image of your Jupyter Notebook code with just a click of a button and visualize your container analysis on the Amazon IoT Analytics console.
Automate container execution - Amazon IoT Analytics lets you automate the execution of containers hosting custom authored analytical code or Jupyter Notebooks to perform continuous analysis. You can schedule execution of your custom analysis on the recurring schedule that best meets the need of your business.
Incremental data capture with customizable time windows - Amazon IoT Analytics enables users to perform analysis on new incremental data captured since the last analysis. You can improve analysis efficiency and lower costs by precisely scanning just your new data. No matter when you ran your last analysis, customizable time windows will capture the new data for you since your last analysis.