Github user mohitgargk commented on the issue:
https://github.com/apache/nifi/pull/3079
> What is the use case for having Parquet files flow through NiFi?
I have a use case where I am writing the parquet files into data lake (in
azure) and also need to write them on a network storage device. The intent here
is keep the parquet in flowfile, so that it can flow into any sink later on. If
we don't have Convert*ToParquet (or similar processors), we would require a hop
(an i/o endpoint) that may not give good performance.
---
