Effect Preservation in Transaction Processing in Rule Triggering Systems

Rules provide an expressive means for implementing database behavior: They cope withchanges and their ramifications. Rules are commonly used for integrity enforcement, i.e., forrepairing database actions in a way that integrity constraints are kept. Yet, Rule TriggeringSystems fall short in enforcing effect preservation, i.e., guaranteeing that repairing events donot undo each other, and in particular, do not undo the original triggering event.A method for enforcement of effect preservation on updates in general rule triggering systemsis suggested. The method derives transactions from rules, and then splits the work betweencompile time and run time. At compile time, a data structure is constructed, that analyzes theexecution sequences of a transaction and computes minimal conditions for effect preservation.The transaction code is augmented with instructions that navigate along the data structure andtest the computed minimal conditions.This method produces minimal effect preserving transactions, and under certain conditions,provides meaningful improvement over the quadratic overhead of pure run time procedures. Fortransactions without loops, the run time overhead is linear in the size of the transaction, andfor general transactions, the run time overhead depends linearly on the length of the executionsequence and the number of loop repetitions. The method is currently being implemented withina traditional database system.