A parallel / distributed database system implemented in Java that was developed during the Spring 2012 semester (my final undergraduate semester) under the guidance of Dr. Lipyeow Lim.  PJDB uses embedded Apache Derby databases for the database nodes and Apache Thrift to handle the communications between database nodes.  The initial system was implemented by Chris Mullins, a master’s student in the University of Hawaii Manoa’s ICS department at the time, and I was tasked with extending the initial system with additional functions such as the CSVPartition function which uses reflection to dynamically load classes to partition data read from a CSV file and distribute the partitioned data to the data nodes.

The source code may be found at: http://code.google.com/p/pjdb/source/browse/ in the trunk directory.
Unfortunately, the project was abandoned following that semester, hence the barren project page and the lack of updates.