A new press release reports, “Elastic N.V., the company behind Elasticsearch and the Elastic Stack, is thrilled to announce that version 7.3 of the Elastic Stack has arrived, and it’s another good one. We’ll touch on some of the release highlights. Be sure to check out the dedicated blogs for each product to dive into all the details of what’s new. If you just can’t wait, version 7.3 is available right now on our Elasticsearch Service — the only hosted Elasticsearch offering to include these new features. Or you can download the Elastic Stack. Now, on to the reason you’re here — here’s what’s new.”
The release goes on, “Say hello to data frames, a new feature that lets you pivot your Elasticsearch data on the fly to create live entity-centric indexes. It’s an exciting feature that opens the door to a new world of analysis, including new machine learning analysis — like outlier detection (which was added as an experimental feature in 7.3), clustering, classification, and more. As with most powerful concepts, this feature is best described with an example. Imagine you want to look for suspicious IP addresses in your web server logs. You might want to look at how many requests were made, the response codes, and the total data transferred for each IP address. Data frames allows you to create a new entity-centric index with a document per unique IP address that tracks each metric of interest — in this case, total requests, count per response status, and sum of bytes transferred. The icing on top is that data frames supports continuous processing, which means that this transformed entity-centric index is automatically updated as new documents are added to the input index.”
Read more at Business Wire.
Image used under license from Shutterstock.com