Working on code that only makes sense against production data... our current "small" dataset is 20TB, our real one is going to land about 150TB. These datasets

Working on code that only makes sense against production data... our current "small" dataset is 20TB, our real one is going to land about 150TB. These datasets are a pain to get locally, and a pain to maintain. Way easier to just have an account on an attached box you can use from anywhere. Another perk of it is using tmux and having a consistent environment you can tap into from work, home, the road, etc.