Load balancing is widely used in parallel forwarding engines and cluster computing environments in high-speed links. Some applications based on packets content, such as IDS and IPS; require keeping network session integrality while balancing the traffic load. That's to say, packets belonging to a network session should be scheduled to the same processing node; otherwise, the semantic integrality of the session will be violated. Traditional adaptive load balancing algorithms are addressed on parallel processing and resolving the performance problem in high-speed links. Based on the related work on load balancing, a load balancing algorithm named adaptive load balancing algorithm based on Minimum Sessions First is proposed in the paper. Simulation results indicate that the scheme achieves significant improvement in session's integrality disruption and has a fairly good load balance both in packets level and bits level. It's a sample algorithm and also can be easily implemented in hardware. The algorithm had been implemented in the high speed network security device designed by National University of Defense Technology.
CITATION STYLE
Chen, Y. J., Lu, X. C., Shi, X. Q., & Sun, Z. G. (2008). Session-oriented adaptive load balancing algorithm. Ruan Jian Xue Bao/Journal of Software, 19(7), 1828–1836. https://doi.org/10.3724/SP.J.1001.2008.01828
Mendeley helps you to discover research relevant for your work.