Reading cloud data and recalculating popularity

(taken from section 4 of the file at https://drive.google.com/file/d/1XdzNaD9XJyMOimvEjzkN1DNjUINdj4p5/view)

Once a day, a Lambda process should ingest all of the data from the S3 bucket uploaded in step #3. It
should record the number of plays per country for each track. The data should be stored on a per-
month, per-country basis. Then, popularity data should be recalculated, but the old popularity index
should be preserved. If there is a significant change in the popularity index, the new popularity data
should be written out to a new file for Comrad instances to ingest. The data could be as follows (doesn’t
have to be CSV):
Artist Names, Track Name, Old Popularity Index, New Popularity Index
Q-tip feat D`Angelo, Believe, 7, 12
There should be a date associated with each popularity index file.