Date of Award:


Document Type:


Degree Name:

Doctor of Philosophy (PhD)


Computer Science


Stephen W. Clyde


Implementing crosscutting concerns for transactions is difficult, even using Aspect-Oriented Programming Languages (AOPLs) such as AspectJ. Many of these challenges arise because the context of a transaction-related crosscutting concern consists of loosely-coupled abstractions like dynamically-generated identifiers, timestamps, and tentative value sets of distributed resources. Current AOPLs do not provide joinpoints and pointcuts for weaving advice into high-level abstractions or contexts, like transaction contexts. Other challenges stem from the essential complexity in the nature of the data, operations on the data, or the volume of data, and accidental complexity comes from the way that the problem is being solved, even using common transaction frameworks. This dissertation describes an extension to AspectJ, called TransJ, with which developers can implement transaction-related crosscutting concerns in cohesive and loosely-coupled aspects. It also presents a preliminary experiment that provides evidence of improvement in reusability without sacrificing the performance of applications requiring essential transactions. This empirical study is conducted using the extended-quality model for transactional application to define measurements on the transaction software systems. This quality model defines three goals: the first relates to code quality (in terms of its reusability); the second to software performance; and the third concerns software development efficiency. Results from this study show that TransJ can improve the reusability while maintaining performance of TransJ applications requiring transaction for all eight areas addressed by the hypotheses: better encapsulation and separation of concern; loose Coupling, higher-cohesion and less tangling; improving obliviousness; preserving the software efficiency; improving extensibility; and hasten the development process.