This message was deleted.
# general
s
This message was deleted.
g
I would go with Option 3: separate Druid cluster with builtin SQL ingest for ingest, using
hdfs
input source 🙂
As to the two options you mentioned, though, I am not really sure what's best, it probably depends on how good you are at managing Druid vs. how good GCP Dataproc is at managing Druid
I am not familiar enough with either one to recommend one over the other