Elastic is an open source search and analytics engine and our number one partner for enterprise search solutions. April 25th, Elastic launched a live virtual event to present their Elastic 7 release. We organised a small pizza-party and watched the live event with some members of the team. In this post, we will give a short overview of what's new in Elastic 7.
Focus on the content
'Focus on the content.’ That was the main thought behind the new Kibana interface. The light, minimalistic design, global navigation header and improved querying and filtering gives users a clear overview of relevant data without any form of clutter. With the help of the new Elastic UI framework, a set of reusable UI components, you can guarantee a consistent user experience with the right look and feel throughout the different dashboards. Want to check Kibana on the go? No problem, the new design is more responsive than ever. And as an icing on the cake for the dark theme enthusiasts, the interface also has the ability to switch to dark mode.
Increased reliability and performance
One of the biggest strengths of Elastic is the possibility to easily scale from one cluster to a larger set of clusters. Cluster coordination is at the basis of that scalability and makes communication and collaboration between different nodes possible. The cluster coordination for Elastic 7 underwent some serious improvements to make it faster, safer, future-proof and easier to use.
- Faster master elections
- Less risks when scaling clusters
- Clearer log status of the nodes
Next to cluster coordination, the Elastic 7 release brings another improvement to the table which brings reliability to the next level, the real-memory circuit breaker. To avoid node overload and failure, Elastic uses circuit breakers to reject requests that would cause memory overflow errors. Currently, these decisions are based on estimations if a request will push the node over its fixed limit.
To make this process more accurate, Elastic 7 introduced real-memory circuit breakers, which rely on the actual memory usage of the Java Virtual Machine instead of estimated calculations to reject the request. This boosts overall search performance and reliability.
Create better search experiences
Search is all about generating relevant results as fast as possible. Elasticsearch was already fast, but to better the search process even more, they added some slight improvements in Elastic 7 to enhance search experiences. You can, for example, limit the depth of your query so that elasticsearch only displays the top results,achieving an even faster response time. To make sure the most relevant results pop up at the top, Elasticsearch refined their function scoring to add ranking to individual records in an easier and more flexible way.
Another new feature is the interval query, a simpler way to find words or phrases in a record which are within a certain reach of each other, mostly used in use cases for patents or legal documents. Next to that, the default shard per index number is now 1 instead of 5, increasing the overall search performance while skipping the unnecessary refresh of ‘search idle’ shards (shards that haven’t had searches for 30 seconds) boosts the indexing performances.
Elastic is evolving rapidly and we believe it’s only the beginning. Interested in what Elastic can do for your business? Don’t hesitate to contact us!