Leslie has a link to Common Lisp Prevalence, an "object prevalence" system for Common Lisp. Object prevalence systems allow you to keep all of your business objects in RAM instead of serializing them back and forth to an RDBMS for every query. With a transaction log, your data is safe even in the case of a system crash.
Here's an article at Advogato explaining Prevayler, an object prevalence system for Java, and here's An introduction to object prevalence at IBM developerWorks.
Object prevalence systems are different than OODBMS because OODBMS usually use some kind of object query language, while object prevalence systems just use your normal programming lanugage operations. As long as your data set is small enough that you won't start paging to disk -- and with systems with many gigabytes of RAM there are many applications for which that's not a concern -- object prevalence systems seem like a great idea.
Feel free to post a comment below. Please see my comment policy.
Formatting Rules (No HTML):