We have a few problems that could be solved using an http endpoint API for hive: 1. There is a delay between when data is uploaded to s3 and when the cron job next runs the import 2. Users must communicate with us to add a dataset to be imported 3. There is no way for users to ad-hoc import datasets One option is https://cwiki.apache.org/confluence/display/Hive/WebHCat A work-around is https://github.com/fbertsch/parquet2hive_server