Grid computing has gained considerable attention in research and industry. High expectations are associated with the approach. However, so far only few papers have been published about the costs caused by Grid computing. In this article we pursue two main goals: to analyze the different types of costs and to determine the total costs of a resource provider. Our approach is based on the discussion of general cost categories that have to be taken into account. We give concrete numbers for the different categories and use these numbers to estimate the costs in two real life Grids: the EGEE project and the Grid of the pharmaceutical company Novartis. A summarizing discussion concludes the paper.

A fundamental problem of grid computing is the communication overhead. One reason of this overhead is the access to remotely stored data. Caching read-only data is a possible alleviation of the problem. In case of grid computing caching can be optimized by using allocation schemes considering the contents of the caches. Possible ways to achieve such an allocation in a grid are the topic of this paper. The paper proposes to use allocation schemes preferring resources with the required data in their caches. In doing so the hit rate of the caches will be increased and as a consequence the average response time of the jobs and the network load will be reduced. Two new possible allocation approaches are discussed and compared with classical allocation schemes. The performance and the costs of the schemes (when applied to large grids) are evaluated using a simulation environment.

No abstract

Component technology has become an important approach to set up distributed applications and systems. Components possess exactly specified or standardized interfaces and explicit context dependencies. They can be purchased at component markets and integrated into the application. The purchase of components, however, might be not the onliest solution in future. If a component requires large resources, an expensive maintenance, or a high degree oftopicality it might be more favourable Iike in everyday Iife to lease a component for a certain time. This can save purchase cost and give the customer the possibility to always use the newest version of a component. The leasing of components requires a particular framework. Beside a contract which defines the conditions ofthe component use, a technical infrastructure is required which ensures the availability ofthe service and the enforcement of the contract agreements. In this paper we present a technical framework for the leasing ofcomponents called Virtual Private Components. We further present an architecture to support the leasing of components and discuss implementation issues. Finally, we present a prototype implementation.