Join Partitions in Spark SQL for better performance

Question

I am new to Spark SQL. I have a question on partition usage during Joins

Assume that there is a table named test1 that saved on 10 partitions (p