Lower bounds for the hub location problem
Management Science
Experiments on traveling salesman heuristics
SODA '90 Proceedings of the first annual ACM-SIAM symposium on Discrete algorithms
Fitness landscapes and memetic algorithm design
New ideas in optimization
An Exact Solution Approach Based on Shortest-Paths for P-Hub Median Problems
INFORMS Journal on Computing
Stochastic Local Search: Foundations & Applications
Stochastic Local Search: Foundations & Applications
An ELSxPath Relinking Hybrid for the Periodic Location-Routing Problem
HM '09 Proceedings of the 6th International Workshop on Hybrid Metaheuristics
Split-Delivery Capacitated Arc-Routing Problem: Lower Bound and Metaheuristic
Transportation Science
Engineering Applications of Artificial Intelligence
Computers and Operations Research
PPSN'10 Proceedings of the 11th international conference on Parallel problem solving from nature: Part II
Engineering Applications of Artificial Intelligence
| Hi-index | 0.00 |
Scalability constitutes a key property in Peer-to-Peer environments. One way to foster this property is the introduction of superpeers, a concept which has gained widespread acceptance in recent years. However, the problem of finding the set of super-peers that minimizes the total communication cost is NP-hard. We present a new heuristic based on Evolutionary Techniques and Local Search to solve this problem. Using actual Internet distance measurements, we demonstrate the savings in total communication cost attainable by such a super-peer topology. Our heuristic can also be applied to the more general Uncapacitated Single Assignment p-Hub Median Problem. The Local Search is then further enhanced by generalized don't look bits. We show that our heuristic is competitive with other heuristics even in this general problem, and present new best solutions for the largest instances in the well known Australia Post data set.