your123 - stock.adobe.com
InfluxDB Cloud overhauls time series database engine
InfluxData updated its InfluxDB Cloud database service with a new engine, new storage and real-time data capabilities, and expanded support for SQL queries.
InfluxData is out today with new capabilities in its InfluxDB Cloud time series database service designed to improve performance and data query capabilities.
The 2012 startup, based in San Francisco, has been expanding the features of its InfluxDB time series database in recent years with capabilities to better handle data coming from IoT devices and event streaming data.
The applications for InfluxDB are diverse, supporting operational and data analytics workloads that include industrial sensors, networking, security, and application monitoring.
Multiple purpose-built databases in the market support time series data, including the open source QuestDB, Amazon Timestream, CrateDB and Timescale. In recent years, the MongoDB document database has steadily added support for time series data as well.
The new update for the InfluxDB Cloud service goes a step further by integrating the open source InfluxDB IOx storage engine built in the Rust programming language, which accelerates time series data storage and query performance.
InfluxDB Cloud also added support for users to access data with SQL queries, in addition to InfluxData's own Flux and InfluxQL query languages.
The IOx technology provides InfluxDB with a columnar storage engine that lets organizations manage time series data.
InfluxData's latest updates boost the usefulness of data that organizations collect and are able to query to with SQL, said RedMonk analyst Stephen O'Grady.
"One of the wider market trends we're seeing with respect to databases is a demand for more versatility. And that fits with InfluxData's direction here as they introduce a columnar storage engine," O'Grady said.
Inside InfluxDB's new time series engine
Stephen O'GradyAnalyst, RedMonk
What InfluxData had essentially done is build a new time series engine for the database, according to Paul Dix, co-founder and CTO.
The engine provides a system for buffering and collecting data that is converted into Apache Parquet files. The Parquet files are kept in object storage with an optimized organization to help accelerate query execution. The data structure is what is known a columnar database, now designed for time series data.
"It's built for real-time workloads, which means you write data and within milliseconds of writing the data is available for queries," Dix said.
InfluxDB bring SQL to time series database
The work to improve InfluxDB involved multiple components written in the open source Rust programming language.
Among the components that InfluxDB uses is the Apache Arrow DataFusion SQL query engine, which is written in Rust. The DataFusion technology lets InfluxDB support SQL queries, a capability it did not have before. Specifically, InfluxDB now supports the PostgreSQL wire protocol for SQL.
That "means you can connect a PostgreSQL client to our cloud offering and it looks like a PostgreSQL database, even though it's not," Dix said.