3 ms·
If you can meet your demands in PG currently, you will probably never need something as heavy as Spark. If PG is currently working for you, then chances are it
by missosoup 7y ago
If you can meet your demands in PG currently, you will probably never need something as heavy as Spark.
If PG is currently working for you, then chances are it'll keep working if you optimise it (partitioning, clustering, etc).
If you do want something else, the particular use cases you described are pretty much exactly what Clickhouse was built for.