InfluxData has released InfluxDB 3.0, a rebuilt version of its time series database aimed at removing long-standing limits on data uniqueness while speeding up queries. The update, now generally available through InfluxDB Cloud Serverless and InfluxDB Cloud Dedicated, is built on a new technical foundation that the company says will anchor future product development.
A rearchitected engine built on Arrow and Rust
InfluxDB 3.0 replaces earlier versions' core infrastructure with Apache Arrow, an open source framework for processing columnar data, and is written in Rust, a language chosen for performance and memory safety. According to InfluxData vice president of product Rick Spencer, the redesign traces back to founder and CTO Paul Dix's goal of building a single database capable of handling the full range of time series workloads - not just metrics, but detailed observational data at scale.
Rachel Stephens, an analyst at RedMonk, said the rework directly targets cardinality, a measure of how many unique values exist in a given database column. "Cardinality is historically one of the biggest limitations of time series databases," she said, noting that high-cardinality data has traditionally made logging and tracing unique events difficult for this class of database.
Performance claims and new capabilities
InfluxData says the new version supports unlimited cardinality alongside high throughput, allowing customers to ingest, transform and analyze hundreds of millions of data points per second. The company also cites real-time query responses up to 100 times faster than prior versions, tenfold improvements in data compression that can lower storage costs, and native support for SQL, letting customers query data using a widely known language rather than a proprietary syntax.
Spencer said the combination of SQL support, faster real-time querying and unlimited cardinality delivers gains in "ingest performance, scalability, resilience and efficiency even as data complexity and cardinality increase."
Broader use cases and what comes next
Stephens said the changes extend InfluxDB's reach into observability tasks such as log and trace analysis, areas it previously handled less effectively, while improving compatibility with the broader Apache Arrow and Apache Parquet ecosystem.
Self-managed versions are planned for later this year: InfluxDB 3.0 Clustered, intended as the successor to the on-premises InfluxDB Enterprise product, and InfluxDB 3.0 Edge. Spencer said the company intends to preserve compatibility with existing Enterprise deployments as customers migrate, with further updates to follow as the 3.0 architecture becomes the base for future releases.