In logstash, the pipeline.workers setting determines how many threads to run to process the filter and output. Do we have similar settings for elasticsearch ingest pipeline?
It seems to work single-threaded and the more processes we create the slower the number of processing of all events, even those that will be in another pipeline, is this correct?
I believe each bulk request is processed in a single thread, but different bulk requests can be processed in parallel. The number of bulk requests that can be processed in parallel is driven by the size of the supporting thread pool, which is generally proportional to the number of available CPU cores on the node.
1 Like
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.