We have just crossed 2 billion items in our datastore. While not 3 billion yet, I expect that to happen later this year.

Too bad Redshift can't handle JSON files: Converting everything will be annoying.

Our idea is to change from JSON on loading to Redshift, continuously. http://www.hapyrus.com/pages/flydata-for-redshift

We do. http://adzerk.com/

When you log every mousedown because the founder misunderstands A/B testing, 3 billion rows is easy to come by. Besides - you're busy changing the world, so you should expect to use the same technology as Facebook and Google.