QoS-Aware Fault Tolerance in Grid Computing through Topology-Aware Replica Placement
Luca Valcarenghi, P. Castoldi · 2006
Quality of service (QoS)-aware fault tolerance is defined as the capability of overcoming both hardware and software failures while maintaining communications QoS guarantees. An integrated fault tolerant scheme combining service replication and path restoration has the potential of providing QoS-aware fault tolerance while maximizing the percentage of recovered connections and minimizing the required service replicas. Previous studies focused on evaluating the optimal performance achievable by the integrated fault tolerant schemes through a mixed integer linear programming (MILP) model. This study concentrate on developing heuristics to place service replicas with topology awareness. The aim of this study is to evaluate whether topology-aware heuristics can approximate MILP optimal solutions