Data networks (2nd ed.)
The design philosophy of the DARPA Internet Protocols
ACM SIGCOMM Computer Communication Review - Special twenty-fifth anniversary issue. Highlights from 25 years of the Computer Communication Review
Introduction to Algorithms
Traffic matrix estimation: existing techniques and new directions
Proceedings of the 2002 conference on Applications, technologies, architectures, and protocols for computer communications
Achieving sub-second IGP convergence in large IP networks
ACM SIGCOMM Computer Communication Review
IGP link weight assignment for operational Tier-1 backbones
IEEE/ACM Transactions on Networking (TON)
Traffic distribution over equal-cost-multi-paths
Computer Networks: The International Journal of Computer and Telecommunications Networking
Joint coverage and link utilization for fast IP local protection
Computer Networks: The International Journal of Computer and Telecommunications Networking
Hi-index | 0.00 |
As a promising approach to improve network reliability, Proactive Failure Recovery (PFR) re-routes data traffic to backup paths without waiting for the completion of routing convergence after a local link failure. However, the diverted traffic may cause congestion on the backup paths if it is not carefully split over multiple paths according to their available capacity. Existing approach assigns new link weights based on links' load and re-calculates the routing paths, which incurs significant computation overhead and is susceptible to route oscillations. In this paper, we propose an efficient scheme for load balancing in PFR. We choose an adequate number of different types of loop-free backup paths for potential failures, and once a failure happens, the affected traffic is diverted to multiple paths in a well balanced manner. We formulate the traffic allocation problem as a tractable linear programming optimization problem, which can be solved iteratively and incrementally. As a result, only the flows affected by the failures are re-allocated to backup paths incrementally without disturbing flows not directly affected by the failures. Simulation results show that our scheme is computationally efficient, can effectively balance link utilization in the network, and can avoid route oscillations.