In various applications such as environmental monitoring and alarm detection, a fusion center or coordination node wishes to obtain a function of the data or measurements via communication networks, rather than the data or measurements themselves. The main focus of this talk is the computation of the frequency histogram or type function over Gaussian networks with orthogonal components, which yields many important statistics such as mean, variance, maximum, minimum, median, and so on. Exact and approximate computation capacity results are established for a class of orthogonal Gaussian networks (with no broadcast and no multiple-access) and for a class of Gaussian networks with multiple-access components (but no broadcast).