This is trivial implementation of the Count Min Sketch (CMS) from Graham Cormode and S. Muthukrishnan as described in (Approximating Data with the Count-Min Data Structure).
CMS is a probabilistic data structure that can be used to keep values associated with items, such as the counts of observations of samples in a stream (e.g. IP addresses, URLs, etc...). The counts are only approximated, i.e. there is a probability for error, but the data structure requires considerably less resources compared to storing all samples and their counts.