Traffic distribution over equal-cost-multi-paths

Authors:
Tat Wing Chim;Kwan L. Yeung;King-Shan Lui
Affiliations:
Department of Electrical and Electronic Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong;Department of Electrical and Electronic Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong;Department of Electrical and Electronic Engineering, The University of Hong Kong, Pokfulam Road, Hong Kong
Venue:
Computer Networks: The International Journal of Computer and Telecommunications Networking
Year:
2005

Citing 3
Cited 3

Cisco IP routing: packet forwarding and intra-domain routing protocols

Cisco IP routing: packet forwarding and intra-domain routing protocols
Internet QoS: Architectures and Mechanisms for Quality of Service

Internet QoS: Architectures and Mechanisms for Quality of Service
Achieving near-optimal traffic engineering solutions for current OSPF/IS-IS networks

IEEE/ACM Transactions on Networking (TON)

A design framework towards the profitable operation of service overlay networks

Computer Networks: The International Journal of Computer and Telecommunications Networking
Load-Balanced IP Fast Failure Recovery

IPOM '08 Proceedings of the 8th IEEE international workshop on IP Operations and Management
Fault tolerant mechanism of bio-inspired adaptive routing system

Proceedings of the 3rd International Conference on Bio-Inspired Models of Network, Information and Computing Sytems

Quantified Score

Hi-index	0.00

Visualization

Abstract

To effectively manage the traffic distribution inside a network, traffic splitting is needed for load sharing over a set of equal-cost-multi-paths (ECMPs). In this paper, a new traffic splitting algorithm, called Table-based Hashing with Reassignments (THR), is proposed. Based on the load sharing statistics collected, THR selectively reassigns some active flows from the over-utilized paths to under-utilized paths. The reassignment process takes place in such a way that the packet out-of-order problem is less severe. Simulations show that THR provides close-to-optimal load balancing performance but suffers about 5% of packets arrived out-of-order. Although additional traffic monitoring function is needed by THR, we show that the extra complexity incurred is marginal.