Data Warehouse: Redshift/Cloud vs. Roll your own
I hope this is the right forum for this topic since it deals with infrastructure.
I'm building a data warehouse for a few sources (using Stitch for ETL) and the specs are not that demanding. We aren't getting more than a million/2 million rows a month. No big data yet. I'm trying to decide between a Redshift cluster vs. rolling out my own Postgres instance.
There are obviously major savings with going the roll-your-own route but I'd be up shit creek if something goes wrong. Redshift feels better since the default cluster has major headroom.
Read responses in arstechnica.com