Packet classification on multiple fields
Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
Deriving traffic demands for operational IP networks: methodology and experience
IEEE/ACM Transactions on Networking (TON)
Traffic Engineering with AIMD in MPLS Networks
PIHSN '02 Proceedings of the 7th IFIP/IEEE International Workshop on Protocols for High Speed Networks
Hop-by-hop routing algorithms for premium traffic
ACM SIGCOMM Computer Communication Review
Fast accurate computation of large-scale IP traffic matrices from link loads
SIGMETRICS '03 Proceedings of the 2003 ACM SIGMETRICS international conference on Measurement and modeling of computer systems
An information-theoretic approach to traffic matrix estimation
Proceedings of the 2003 conference on Applications, technologies, architectures, and protocols for computer communications
BRITE: An Approach to Universal Topology Generation
MASCOTS '01 Proceedings of the Ninth International Symposium in Modeling, Analysis and Simulation of Computer and Telecommunication Systems
Achieving near-optimal traffic engineering solutions for current OSPF/IS-IS networks
IEEE/ACM Transactions on Networking (TON)
Achieving Fast BGP Reroute with Traffic Engineering Using Multiple Routing Planes
IPOM '08 Proceedings of the 8th IEEE international workshop on IP Operations and Management
Adaptive multi-topology IGP based traffic engineering with near-optimal network performance
NETWORKING'08 Proceedings of the 7th international IFIP-TC6 networking conference on AdHoc and sensor networks, wireless networks, next generation internet
Towards decentralized and adaptive network resource management
Proceedings of the 7th International Conference on Network and Services Management
Hi-index | 0.01 |
This paper proposes and evaluates a novel, edge-based approach, which we call the k-set Traffic Engineering (TE) method, to perform traffic engineering in OSPF networks by partitioning traffic into uneven k-traffic sets. The traffic partitioning and splitting takes place only at network edges, leaving the core simple. We theoretically prove that if k is large enough, the k-set TE method achieves the general optimal traffic engineering where full-mesh overlaying and arbitrary traffic splitting, such as in MPLS, have to be used. We give an upper bound of the smallest k that achieves such a general optimum. In addition, we provide a constant worst case performance bound if k is smaller than the optimal k. Finding the optimal traffic splitting and routing for a given k is NP-hard. Therefore, we present a heuristic algorithm to handle the problem. The performance of the k-set TE method together with the proposed heuristic algorithm is evaluated by simulation. The results confirm that a fairly small k (2 or 4) can achieve good near-optimal traffic engineering. Overall, the k-set TE method provides a simple and efficient solution to achieve load balancing in OSPF networks. It follows the ''smart edge, simple core'' design rule of the Internet. It is also able to keep ''the same path for the same flow,'' which is desirable and beneficial to TCP applications.