Amazon Web Services Inc. today said it’s extending the machine learning capabilities of its Amazon Kinesis Data Analytics tool to detect so-called “hotspots” in streaming data, enabling it to provide greater insights than before.

Launched in 2016, Amazon Kinesis Data Analytics is the public cloud giant’s real-time processing engine for streaming data. The service lets users query data streams in the SQL programming language, and then outputs the results of those queries to Kinesis Data Firehose, Kinesis Data Streams or an AWS Lambda function. It’s basically a tool for obtaining actionable insights from streaming data in real time, and can also be used to help build streaming applications using SQL without having to learn any new programming skills.

Amazon’s senior technical evangelist Randall Hunt said in a blog post that the new HOTSPOTS function in Kinesis Data Analytics is used to help identify what he calls “dense regions” or “hotspots” in data streams, without the need to build and train a machine learning model first. These hotspots are essentially regions of activity in data streams that are significantly higher than the norm, Hunt explained. The HOTSPOTS features allows users to programmatically take action on them by streaming the hotspots to a Kinesis Data stream, to a Firehose delivery stream, or by invoking a AWS Lambda function.

In a nutshell, Amazon is offering a more effective way to spot anomalies in data and take action or grab insights on them. Hunt offers a couple of scenarios where the HOTSPOTS feature could be useful. For example, it could be used by the operator of a ride-hailing program to obtain information about severe road conditions, traffic congestion and so on. Data center operators could use HOTSPOTS to identify when a batch of servers begin to overheat, something that could indicate problems with heating, ventilation or air conditioning.

HOTSPOTS is available for Amazon Kinesis Data Analytics in all AWS regions as of today.

Image: Thomas Cloer/Flickr