In some examples, a plurality of multi-dimensional data samples representing respective behaviors of entities in a computing environment are sorted, where the sorting is based on values of dimensions of each respective multi-dimensional data sample. For a given multi-dimensional data sample, a subset of the plurality of multi-dimensional data samples is selected based on the sorting. An anomaly indication is computed for the given multi-dimensional data sample based on applying a function on the multi-dimensional data samples in the subset. It is determined whether the given multi-dimensional data sample represents an anomalous entity in the computing environment based on the computed anomaly indication.