In some examples, a method can include receiving, at a network monitor, discrete side information from a first server at a first rack regarding a data flow between the first server and a second server at a rack other than the first rack. The discrete side information can, for example, include an indicator determined by the first server that indicates whether the data flow satisfies a reference criteria. The method can further include performing, with the network monitor, a network inference process partly based on the received discrete side information.