Comparison of Hash Strategies for Flow-Based Load Balancing

10.7903/ijecs.1346

Surasak Sanguanpong；Witsarut Pittayapitak；Kasom Koht-Arsa

Hashing, Hash Functions ； Flow-based ； Load Balancing

International Journal of Electronic Commerce Studies

6卷2期（2015 / 12 / 01）

259 - 268

英文

Hashing is powerful tool and widely used for flow-based load balancing schemes in parallel processing. In this paper, we analyze and compare computing overhead and load dispersion characteristics of hash strategies using XOR and CRC operations under four hashing key schemes (from 2-tuple to 5-tuple). We conduct experiments with real-life 24-hour campus network traffic. The results show that XOR32 has the lowest computing overhead among all hash function groups. Moreover, XOR32 with 4-tuple and XOR32 with 5-tuple are the two outstanding strategies that provide very good uniform distribution of traffic across multiple links, thus achieving better load balancing for flow-based applications.

1. Ayuso, P. N.,Gasca, R. M.,Lefevre, L.(2009).Demystifying cluster-based fault-tolerant firewalls.Internet Computing,13(6),31-38.
2. Cao, Z.,Wang, Z.,Zegura, E.(2000).Performance of hashing-based schemes for Internet load balancing.Proceedings of the Annual Joint Conference of the IEEE Computer and Communications Societies (vol.1),Tel Aviv:
3. Detal, G.,Paasch, C.,Linden, S.,Mal’rindol, P.,Avoine, G.,Bonaventure, O.(2013).Revisiting flow-based load balancing: Stateless path selection in data center networks.Computer Networks,57(5),1204-1216.
4. Doering, A.,Waldvogel, M.(2004).Fast and flexible CRC calculation.Electronics Letters,40(1),10-11.
5. Feldman, R.,Valdez-Flores, C.(2010).Feldman, and C. Valdez-Flores, Applied probability and stochastic processes.New York:Thomson Publishing.
6. Guang, C.,Wei, Z.,Jian, G.(2008).XOR hashing algorithms to measured flows at the high-speed link.Proceedings of the International Conference on Future Generation Communication and Networking,Hainan Island:
7. Jiang, H.,Jiang, H.,Iyengar, A.,Nahum, E.,Segmuller, W.,Tantawi, A.,Wright, C.(2012).Design, implementation, and performance of a load balancer for SIP server clusters.IEEE Transactions on Networking,20(4),1190-1202.
8. Koht-arsa, K.,Sanguanpong, S.(2012).High availability and scalable parallel stateful firewall design.International Conference on Internet Studies,Bangkok:
9. Koht-arsa, K.,Sanguanpong, S.(2013).A centralized state repository approach to highly scalable and high-availability parallel firewall.Journal of Computers,8(7),1664-1676.
10. Kounavis, M.E.,Berry, F.L.(2005).A systematic approach to building high performance software-based CRC generators.Proceedings of the 10th IEEE Symposium on Computers and Communications,Washington, DC, USA:
11. Molina, M.,Niccolini, S.,Duffield, N.G.(2005).A comparative experimental study of hash functions applied to packet sampling.19th International Teletraffic Congress,Beijing:
12. Nam-Uk, K.,Jung, S.,Chung, T.(2011).An efficient hash-based load balancing scheme to support parallel NIDS.Lecture Notes in Computer Science,6782,537-549.
13. Prabhavat, S.,Nishiyama, H.,Ansari, N.,Kato, N.(2011).On load distribution over multipath networks.IEEE Communications Surveys and Tutorials,14(3),662-680.
14. Xiong, B.,Yang, K.,Li, F.,Chen, X.,Zhang, J.,Tang, Q.,Luo, Y.(2014).The impact of bitwise operators on hash uniformity in network packet processing.International Journal of Communication Systems,27(11),3158-3184.