Performance evaluation of communication networks for parallel and distributed systems

摘要

This special issue of Parallel Computing brings together eight research papers reflecting some of the latest original achievements in the field of performance evaluation of communication networks for parallel and distributed systems. Some of these papers have been selected as extended and revised versions, subject to peer review, from those originally accepted for presentation at the International Workshop on Performance Evaluation of Networks for Parallel, Cluster and Grid Computing Systems (PEN-PCGCS'05), which was held in conjunction with the 2005 International Conference on Parallel processing (ICPP'05), Oslo, Norway, June 14-17, 2005. Each paper has been reviewed by at least two or three referees drawn from the Programme Committee and other experts worldwide. The selected papers address a range of issues encountered in the performance evaluation of networks designed for parallel, cluster, and grid systems, and their contributions are outlined below. In the first paper, Gatani, Re, and Gaglio suggest a distributed heuristic solution for the Steiner tree problem, allowing the construction of effective distribution trees using a coordination protocol among the network nodes. The authors also propose a new distributed technique for dynamically updating the multicast tree. Their suggested approach has been extensively tested in simulation and on experimental networks. The obtained results show that the new algorithm can perform as well as the centralised version, providing good levels of convergence time and communication complexity.