Hierarchical end-to-end service recovery

Failed networks, for example MPLS, can cause signaling storms the size of which can grow dramatically with network size. This paper presents a new scalable fault notification protocol that reduces the size of this storm. The protocol causes failure notification signals to travel vertically up and down a network hierarchy instead of horizontally along the service routes. This reduces the signaling required for longer paths and provides signaling scalability for larger network sizes. The protocol also alleviates the number of fault notification signals by hierarchically aggregating them. Aggregation is based on path management information maintained in the network hierarchy. We show how the protocol reduces the size of the signaling storm analytically and with simulation. However, this comes at the price of increased storage overheads when compared with existing restoration techniques.