Curious, how long it will take to create index on single node for a such large table?

That's about 1/250th of our working set size. :)

Of course, that's in a Spark cluster running massively parallel queries and not on a single (however large) node.

You did not ask that question. I assume you knew before that PostgreSQL is not the tool for your workload.

I didn’t ask a question. I love PostgreSQL for small relational stuff (where “small” is up to a few TB) and use it any time it’s appropriate. I wouldn’t classify it as a Big Data store, though.

That’s no criticism at all. I wouldn’t use Spark for a transactional DB because it’s not the best tool for the job. Same for PostgreSQL when the dataset grows far beyond what a single instance can reasonably process.