how exactly is this solution easier than putting the very Parquet files on a classic filesystem. Why does the easy solution require an amazon-subscription?
This is adjacent to "why would I need EC2 when I can serve from my laptop?"
In terms of maturity of solution and diversity of downstream applications you'll go much further with BigQuery/Athena (at comically low cost) for this amount of data than some cobbled together "local" solution.
I thoroughly agree with the author but the comments in this thread are an indication of people who haven't actually had to do meaningful ongoing work with modest amounts of data if they're suggesting just storing it as plain text on personal devices.
I'm not advocating for complicated or expensive solutions here, BigQuery and Athena are very low complexity compared to any of the Hadoop et-al tooling (yes Athena is Trino is in the family, but it is managed and dirt cheap).
