Is this use case suitable for Spark (iterations)?

Question

I have approximately a dataset (df1) of 130 million rows. The dataset has approximately 90 columns but by this case we only need four of them, so since the