Fluentd

Fluentd is an open-source data collector for unifying log aggregation — it collects logs from many sources, filters and transforms them, then routes them to one or more destinations (Elasticsearch, S3, BigQuery, etc.) through a single, consistent pipeline.

Its plugin ecosystem (500+ community plugins) covers most common input sources and output destinations, and it's a CNCF graduated project, making it a common default choice for centralizing logs in a Kubernetes or microservices environment.

Often compared to Logstash; Fluentd is generally lighter on memory, which is part of why it's the more common choice for a per-node log collector in Kubernetes clusters specifically.