Best practice to handle out-of-memory data

I have been looking for a replacement of MonetDBLite for my own workflow.
I'm working with ~ 50M rows databases, and was not very satisfied/reassured with MonetDBLite as I'm having some corruption in my DB from time to time.

Looking for an alternative, I did run some benchmarks that can be summed up in this plot :

So, for me, although I'm not that much confortable using MonetDBLite, I'll keep using it as there seems to be no free alternative for running "real-time" analysis (I'm making plots with shiny out of this DB) with such quantity of data.

8 Likes