Fluent Bit can read from
local files and network devices, and can scrape metrics in the Prometheus
format from your server.
All events are automatically tagged to determine filtering, routing, parsing, modification and output rules.
Filters can modify data by calling
an API (E.g. Kubernetes),
remove extraneous fields,
or add values.
Built-in reliability means if you hit
a network or server outage
you will be able to resume
from where you left off
without data loss.
Fluent Bit can send data to a
multitude of locations,
including popular destinations
like Splunk, Elasticsearch,
OpenSearch, Kafka, and more.
It has been designed as a lightweight solution with high performance in mind. From a design perspective, it’s fully asynchronous (event-driven) and take the most of the operating systems API for performance and reliability.
FTS — Fluent Typed Scripting adds programmable policies to input and output processors. Work with typed log values, metric families, spans, and OpenTelemetry context to redact fields, normalize data, and reduce noise before forwarding it.
FTS is included in Fluent Bit 5.2. See a policy in action and get started.
In 2014, the Fluentd team at Treasure Data began to see the need for a more lightweight log processor to be used in resource-constrained environments like embedded Linux and gateways. The objective was to provide all the speed, scale, and flexibility of Fluentd in a smaller, more efficient footprint. The result was Fluent Bit.
While Fluent Bit did gain rapid adoption in embedded environments, its lightweight, efficient design also made it attractive to those working across the cloud. Features to support more inputs, filters, and outputs were added, and Fluent Bit quickly became the industry standard unified logging layer across all cloud and containerized environments.
Fluent Bit has been deployed over a billion times and is trusted by some of the world’s largest and most complex organizations.