Spark SQL: Partitions and Sizes

Data fetching parallelisation in Apache Spark through Spark SQL / JDBC.
PartitionColumn, lowerBound, upperBound, numPartitions and a lot of other parameters: how do they work?