llijiajun / NDV_Estimation_in_distributed_environment

Sampling-based estimation of the number of distinct values in distributed environment

Environment

Simulated Experiment

• Ubuntu
• C++ 11
• GCC 4.8

Experiments on Spark

• Scala
• Python 3.7
• JDK
• Hadoop 3.3
• Spark 3.1

Preparation

Generate sampling data from Poisson Distribution and Zipfian Distribution.

python genpoi.py
python genzipf.py