Optimizing the Replication of Multi-Quality Web Applications Using Aco and Wolf

Since its introduction, the Ant Colony Optimization (ACO) meta-heuristic has been successfully applied to a wide range of combinatorial problems. This thesis presents the adaptation of ACO to a new NP-hard problem involving the replication of multi-quality database-driven web applications (DAs) by a large application service provider (ASP). This problem is a special case of the generalized assignment problem (GAP) which occurs in many military contexts such as logistics planning, air crew scheduling, and communications network management. The ASP must assign DA replicas to its network of heterogeneous servers so that user demand is satisfied at the desired quality level and replica update loads are minimized. The ACO algorithm proposed, AntDA, for solving the ASP's replication problem is novel in several respects: ants traverse a bipartite graph in both directions as they construct solutions, pheromone is used for traversing from one side of the bipartite graph to the other and back again, heuristic edge values change as ants construct solutions, and ants may sometimes produce infeasible solutions. Although experiments show that AntDA outperforms several other solution methods, there was room for improvement in the convergence rates of the ants in finding better solutions.