Parquet without Hadooplex

Has anyone come up with a “simple” process to convert files to Parquet format without the need to use a Hadooplex and without going through HDFS.

I would like to read a text file from S3 and convert it to Parquet back onto S3.

I am not aware of a method to do so at the moment; however, Parquet <-> S3 has been an increasingly requested feature recently and we are actively investigating implementation options, though we don’t have a timeline for release at the moment.

Shayne Hodge

Realize this is an old thread, but just wanted to update. In our release last month (Summer), this functionality was added.