InfluxData today added a dashboard along with single sign-on capabilities and a set of application programming interfaces (APIs) for managing its time series database.
In addition, InfluxDB Clustered, an edition of the database that can be deployed on Kubernetes clusters, is now also generally available.
InfluxDB CEO Evan Kaplan said IT teams are now implementing time-series databases across a wider range of use cases, as precisely when data was created, processed and analyzed becomes a more critical element of application. For example, when exactly high cardinality data was created now plays a major role in the training of large language models (LLMs) used in generative artificial intelligence (AI) applications, he noted.
Additionally, more data at the network edge is now being processed and analyzed in real time at the point where it is created and consumed, added Kaplan.
InfluxDB provides both an edition of its database that is managed in the cloud as a service, and an on-premises edition that IT teams can deploy and manage themselves on Kubernetes clusters. Both editions on based on version 3.0 of InfluxDB that is now based on a columnar engine, dubbed IOx, that leverages the open source Apache Arrow memory format and is written in the Rust programming language.
That foundation makes it possible to continuously ingest, transform and analyze hundreds of millions of time series data points per second. At the same time, InfluxDB takes advantage of high compression object storage to reduce the total cost of storing all that data. It also provides interoperability with Open Data Architecture (ODA) to integrate with data lakes based on open source platforms such as DataFusion, Flight SQL and Parquet that are being advanced by the Apache Software Foundation.
Version 3.0 also takes advantage of Kubernetes to enable IT teams to independently scale various tasks as needed. For example, more compute resources can be specifically allocated to processing queries or, conversely, ingesting data.
It’s not clear to what degree time-series databases are being more widely used, but the number of applications that process data in real time in, for example, operational technology (OT) environments continues to increase. At the same time, the number of platforms that rely on time-series databases to go beyond monitoring metrics to also observe IT environments by analyzing logs and traces in near real time is also increasing.
There have, of course, never been as many database options for processing and analyzing data as there are today. Each additional type of database added to an IT environment brings with it a unique set of management challenges. The challenge IT teams encounter is finding and retaining the IT expertise needed to manage multiple types of databases that are now being used to drive a wider range of classes of applications.
The days when the bulk of applications were based on batch-oriented process invoking a relational database are over. In its place is a much more complex IT environment that requires a level of data management and engineering expertise that for many organizations is still hard to find and retain.
