Flow-Level Multipath Load Balancing in MPLS Network

Zenghua ZHAO
Yantai SHU
Lianfang ZHANG
Oliver YANG

IEICE TRANSACTIONS on Communications   Vol.E88-B    No.5    pp.2015-2022
Publication Date: 2005/05/01
Online ISSN: 
DOI: 10.1093/ietcom/e88-b.5.2015
Print ISSN: 0916-8516
Type of Manuscript: PAPER
Category: Network
flow level model,  multipath load balancing,  MPLS network,  

Full Text: PDF>>
Buy this Article

Multi-Protocol Label Switching (MPLS) can efficiently support the explicit routes setup by the use of Label Switched Paths (LSPs) between an ingress Label Switched Router (LSR) and an egress LSR. Hence it is possible to distribute the network traffic among several paths to achieve load balancing, thus improving the network utilization, and minimizing the congestion. The packet-level traffic characteristics in the Internet is so complex that it is natural to do traffic engineering (TE) and control at the flow level. The emerging Multi-Protocol Label Switching (MPLS) has introduced an attractive solution to TE in IP networks. The main objective of this paper is to balance traffic at the flow level among the parallel Label Switched Paths (LSPs) in MPLS networks. We introduce a multipath load-balancing model at the flow level. In this model, each LSP is modeled as an M/G/1 processor-sharing queue. The load-balancing problem is then considered as an optimization problem. Based on the analysis of the model, we propose a heuristic but efficient mechanism that can make good use of the traffic characteristics at the flow level. Packet disorder is avoided effectively by dispatching packets belonging to one flow to the same path. This mechanism only need to be implemented in the ingress LSRs and the egress LSRs, while the intermediate LSRs only forward the packets. Apart from discussing the traffic allocation granularity, and the implementation issues in details, we have also performed extensive simulations using NS-2 with MPLS modules. The simulation results show that the load through the network is well balanced so that the network throughput is improved and the delay is decreased efficiently.