What do you like best?
I love how easy it is to use to store columnar data. Once you learn the details, it makes Hadoop-use a breeze. Column-store data has many benefits, and Parquet is such a help.
What do you dislike?
There is definitely a learning curve with the environment, but it is minimal. There honestly is not much I dislike about it.
Recommendations to others considering the product:
I would definitely recommend Apache Parquet if you are considering using columnar-store data!
What problems are you solving with the product? What benefits have you realized?
I had to gather raw data and consolidate it in a way to run statistical analysis and machine learning on it. Apache Parquet made my job a lot easier. This data analysis provided a huge step in the completion of the project.