I don't think there is a much difference if you're not processing data on the fly. You'd find both the languages doing fair on that point. Like say Python can process both server side and desktop side data pretty quickly.

R too have shown it's performance on offline data. However I am skeptical on performance of R language on the server side. e.g. Reading and saving data files in R often have different performance speed for Python and R. I have found server side, python performing better.

Depending on how quickly you want to process the data, it'd be different for both languages. My preference in such case is for Python.

I prefer Python because I find the language itself more logical and consistent (and it's more general, so more worthwhile to learn if you also do other programming tasks).

But Python and R are the two common options for such tasks, and both are good choices without too much difference functionally.

It's not much data you can use R pretty safely and would have no problem ever regarding speed. Just use data.table package from R

Today data science experts have an extremely wide range of various tools available which they can use to cope with literally any issue they might face. However, there are two programming languages, which are commonly acknowledged as the best tools for data science projects: R and Python.