ISENGARD: an Infrastructure for Supporting e-Science and Grid Application Development

Grid computing facilitates the aggregation and coordination of resources that are distributed across multiple administrative domains. The combined infrastructure provides data storage and processing capacity beyond that of individual organizations; and enables large-scale, complex, and multi-domain e-Science experiments. Grids are inherently distributed and heterogeneous; and this poses challenges such as data access and authentication methods. To alleviate these concerns, the grid community has developed specific middleware that provides a uniform set of services for security, information, and management. Grid middleware is used to hide the heterogeneous nature and provide users with a homogenous and seamless environment by implementing a set of standardized interfaces to a variety of services. Writing, deploying, and testing grid applications over highly heterogeneous and distributed resources is complex and challenging. The process requires grid-enabled programming tools that can handle the complexity and scale of the infrastructure. While a large amount of research has been undertaken into grid middleware, little work has been directed specifically at the area of grid application development tools. As a result, there is a lack of tools for building, testing, managing, and deploying grid-enabled software; and this impedes the uptake of grid computing. This thesis presents the design, implementation, and demonstration of ISENGARD, an infrastructure for supporting e-Science and grid application development. ISENGARD consists of three distinct software components with well-defined APIs and bindings. Specifically, they include: an IDE framework which can be incorporated into existing IDEs for developing grid vii applications; a middleware-independent framework that provides high-level development services; and a modular framework in the form of a grid debugging service for finding and identifying software bugs in grid applications.