Select sampling methods for probability samples using large data sets. This includes spatially balanced sampling in multi-dimensional spaces with any prescribed inclusion probabilities. All implementations are written in C with efficient data structures such as k-d trees that easily scale to several million rows on a modern desktop computer.