We address these two shortcomings by developing a family of new algorithms for the computation of domain-domain interactions based on the idea of bounding shapes, which are used to prune the search space. The best of the algorithms improves on the old PSIMAP algorithm by a factor of 60 on the PDB. Additionally, the algorithms allow a distributed computation, which we carry out on a farm of 80 Linux PCs. Overall, the new algorithms reduce the computation at atomic level from months to 20 min. The combination of pruning and distribution makes the new algorithm scalable and sustainable even with the superlinear growth in PDB.

SUMMARYResource allocation is an important aspect of Grid computing. Over the past few years, various systems have been developed which use market mechanisms to allocate resources. However, the performance of such policies has not been sufficiently studied. In this paper, we investigate under which circumstances market-based resource allocation by continuous double auctions and by the proportional share protocol, respectively, outperforms a conventional round-robin approach. We develop a model for clients, servers and the market, and present simulation results. Factors which are studied include the amount of load in the system, the number of resources, different degrees of resource heterogeneity, and communication delays.

We review mobile agents in the context of distributed object computing and parallel processing. We compare these three paradigms qualitatively. For a quantitative comparison of RMI and Voyager as mobile agent platforms, we identify distributed information processing with flexible load balancing as a convincing application to evaluate the two platforms.

No abstract

No abstract