Appendix B. Reviewing relational databases
Relational databases have been an integral part of enterprise development for a few decades now. The fact that these business data storage mainstays are backed by their own body of mathematical theory (relational algebra) speaks to the elegance and robustness of this mature technology. E. F. Codd first introduced the theory of relational databases in 1970 while working at IBM. This groundbreaking research eventually led to the creation of today’s database products, including IBM’s own highly successful DB2 database. Oracle is the most popular database in existence today, in vibrant competition with products like Microsoft SQL Server, Sybase, MySQL, and many others, in addition to IBM’s DB2. Fundamentally, relational databases store and organize related data into a hierarchy of schemas, tables, columns, and rows.
Other types of databases exist, including flat-file, hierarchical, network, and object-oriented databases. Each of these is worthy of study on its own merits. However, the EJB 3 specification only supports relational databases, and that will be where we draw the line with regard to the database discussions in this book. The focus of this appendix is to briefly discuss each of the relational concepts.