Wang and Jie present a computational environment for scientists and other grid computer users that is simple enough to allow them to take advantage of its power, yet shields them from its intricacies. Their approach consists of a software layer to be put on top of existing grid computing applications that acts as a user interface and virtualizes the entire grid environment.
Grid computing distributes the burden of processing single computation-intensive applications among many geographically dispersed computers, harnessing their collective computational power and bringing them together to solve problems that would otherwise require far more powerful and complex machines. Over the years, this approach has proven to be effective and powerful, especially in solving scientific problems. Yet, the computational skills needed to take advantage of this discipline are not trivial and often add an extra layer of complexity to the scientific problem being investigated.
The paper presents the overlay philosophy and the European academic consortium established to carry on the project. Then, it describes the test environment where these concepts have actually been implemented, complete with some code snippets from the test environment itself. Finally, it discusses future evolution of the project, as well as related work by other groups.
The paper will be useful to grid computing researchers who want to stay current with the latest advances in the field, as well as those who need an introduction to the field of grid computing.