GAUGE: Grid automation and generative environment using domain engineering and domain modeling for drafting applications for the grid.

摘要

This dissertation describes an alternative model for Grid applications development. Grid applications not only need to adapt to the evolving standards but they also need to adjust to a dynamic environment in which resources can be added and removed on demand. Because of these reasons, issues like rapid prototyping and maintainability must be at the core of the applications development model. The approach presented in this dissertation aims at hiding unnecessary accidental complexities met when developing Grid applications. To realize this goal, the work focuses on constructing the Grid Automation and Generative Environment (GAUGE) using concepts of domain engineering and domain-specific modeling which are modern software engineering methods for automating the development of software. GAUGE improves both the development of new Grid applications as well as their maintainability. By focusing on the problem at the modeling level rather than the specific technology used for a solution, GAUGE can protect Grid specialists and users from technology obsolescence. That is, GAUGE models are not tied to any one solution and can easily adapt to the next generation's new technology. Experimental evaluation of this work is performed by comparing it with other approaches and by illustrating the effort saved by using this tool.;A major tenet of this dissertation is that for the tools to be effective, it is of paramount importance that they should rely on modern software engineering practices rather than ad-hoc methods. This allows for increased software quality as well as improvement of the development process for such tools. Tools that can assist programmers during the development process are imperative to simplify the way in which we program. Furthermore, tools that empower end-users to take full advantage of the Grid infrastructure without relying on support teams are essential to fulfill the original Grid vision of computing power as a utility accessible to all.