I am looking for Filebeat performance benchmark numbers and how it compares other shippers like FluentD.
Any pointers would help.
In our experience Filebeat is rarely the bottleneck. It's either the next stage (Logstash / Elasticsearch / Kafka, etc.), or the network, or the disk throughput.
If you want, you can benchmark Filebeat in isolation by using the file output, or the console output redirected to /dev/null. FIlebeat prints some stats every 30 seconds. The number you will get depends on your hardware, we see above 100K per second on good HW. But as Filebeat never runs in isolation, this will be just the upper limit of what you can get in practice.
Thanks.
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.