What is MPEG 7

The strength of grid computing, namely being able to aggregate distributed computing and data storage capacities for solving larger, more complex problems, can currently only be partially exploited because of difficulties in accessing grid infrastructures, lack of usability, security or legal concerns, and missing performance guarantees or billing mechanisms. In this chapter, we present application scenarios where some of those weaknesses are overcome by presenting the user with transparent, intuitive, location independent access to grid resources using techniques from mobile and pervasive computing. Two approaches are suggested. In the implicit approach, user needs and context information are to be captured by means of smart devices and objects trying to perceive the jobs and tasks users want or need to delegate for computation in the grid. In the user-controlled approach, users explicitly provide meta-information on the type of tasks to be solved and the underlying supporting infrastructure provides the most suitable type of applications as well as mechanisms for returning the results to the user. Both approaches are briefly described in the chapter by means of a specific sample scenario in the field of image analysis. Other application areas for mobile and ubiquitous grids based on our experiences gained in the Austrian grid project are also presented in the paper.