Deep dive into the InfluxDB 3 Core open source release
Thursday, May 8
2:00 PM
30 mins
Main Stage
About this session
Four years in the making, InfluxDB 3 Core is the new open source release of InfluxDB 3, a time series and analytic event database built with Apache Arrow, DataFusion, and Parquet. In this session, InfluxDB creator Paul Dix will cover InfluxDB Core’s diskless design, which gives it the ability to operate solely with object storage (in addition to a regular local disk). He will also discuss the new API, including SQL capabilities, Last Value Cache, Distinct Value Cache; the compatibility differences with InfluxQL and previous InfluxDB versions; and the integrated Python VM that acts as an embedded processing engine. For those familiar with the TICK stack, this makes InfluxDB Core capable of filling roles previously performed by Telegraf and Kapacitor. He’ll demo the plugin API with examples for building plugins triggered on WAL writes, Parquet file persists, scheduled tasks, and custom request endpoints, and go into the details of configuring Grafana to work with InfluxDB Core.
