An easy way to think about Zipfian distributions is something like "the most popular 30% of your data set is queried 70% of the time". Its basically the temporal locality property that a lot caching systems exploit. For…
An easy way to think about Zipfian distributions is something like "the most popular 30% of your data set is queried 70% of the time". Its basically the temporal locality property that a lot caching systems exploit. For…