Why this matters¶
At some point your work outgrows your laptop — a simulation that runs for days, a dataset that will not fit in memory, or a model you cannot train overnight. Waiting is not a strategy, and neither is buying a bigger laptop.
Renting compute by the hour is cheaper and less intimidating than most people assume, and TU/e offers options well before you need a credit card. The genuinely hard part is not the machine but the logistics — getting your code and data there and your results back — which is precisely the problem containers solve.
From the session¶
Session notes, slides, and worked examples to follow.