Anscombe's quartet shows us that datasets with the same statistical characteristics (e.g., mean, variance), may in fact consist of very different raw data values. Visualizing these datasets helps us recognize their differences.

Shapes are pre-attentively processed.

What are some compelling reasons for why the use of colors in data visualization is effective:

Visualization can help with data exploration

HBase is column-oriented database.

A machine in a Hadoop cluster could serve as both a Mapper and Reducer.

DASK must run on a many-core server, and cannot run on commodity machines like laptop computers.

DASK supports parallel computing in Python.