HyperLogLog-based accumulator for Apache Spark
A custom AccumulatorV2 implementation that uses HyperLogLogPlus to count unique elements seen. It provides a cardinality estimation of the observed elements.
This sample application is meant to be deployed local, using the sbt publishing function
sbt publishLocal