Hi everyone, I've recently got curious of what are benefits/drawbacks for Beam on X vs X, where X is relevant runner (Spark, Hadoop, etc).
I wonder, if anyone did similar research already and might have some documents/tables/references available? Sample topics of curiosity: * performance of similar pipelines * ease of development * debugability * extra/missing functionality (we have capability matrix <https://beam.apache.org/documentation/runners/capability-matrix/>) * other topics? Regards, --Mikhail