HLB: Toward Load-Aware Load Balancing

Zhiyuan Yao (1, 2) , Yoann Desmouceaux (1) , Juan-Antonio Cordero-Fuertes (3) , Mark Townsley (1) , Thomas Clausen (3)

1 Cisco Systems France
2 X-DEP-INFO - Département d'informatique de l'École polytechnique
3 LIX - Laboratoire d'informatique de l'École polytechnique [Palaiseau]

The purpose of network load balancers is to optimize quality of service to the users of a set of servers-basically, to improve response times and to reducing computing resources - by properly distributing workloads. This paper proposes a distributed, application-agnostic, Hybrid Load Balancer (HLB) that-without explicit monitoring or signaling-infers server occupancies and processing speeds, which allows making optimised workload placement decisions. This approach is evaluated both through simulations and extensive experiments, including synthetic workloads and Wikipedia replays on a real-world testbed. Results show significant performance gains, in terms of both response time and system utilisation, when compared to existing load-balancing algorithms.

Mots clés: load-balancing; cloud and distributed computing; performance evaluation

Domaines: Réseaux et télécommunications [cs.NI]; Calcul parallèle, distribué et partagé [cs.DC]

Zhiyuan Yao, Yoann Desmouceaux, Juan-Antonio Cordero-Fuertes, Mark Townsley, Thomas Clausen. HLB: Toward Load-Aware Load Balancing. IEEE/ACM Transactions on Networking, 2022, pp.1-16. ⟨10.1109/TNET.2022.3177163⟩. ⟨hal-03689859⟩