I work on hierarchical agglomerative clustering on large amounts of multidimensional vectors, and I noticed that the biggest bottleneck is the construction of the distance matri
