Open AccessAn INT-based Load Balancing Mechanism for Cloud Datacenters
Author(s) -
Zixi Cui,
Yuxiang Hu,
Saifeng Hou
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1738/1/012071
Subject(s) - computer science , cloud computing , scalability , load balancing (electrical power) , distributed computing , computer network , overlay network , clos network , virtual machine , virtualization , virtual network , enhanced data rates for gsm evolution , software deployment , operating system , the internet , telecommunications , geometry , mathematics , grid
In this work, we propose Closer, a scalable congestion-aware load balancer for cloud datacenters. Closer complies with the evaluation of technology including the deployment of regular Clos topologies, overlays for network virtualization, and virtual machine (VM) clusters. Leveraging In-band Network Telemetry (INT) to obtain precise link load information, we design a simple but efficient algorithm that implements weighted ECMP at the edge of fabric, which enables Closer to proactively map the flows to the appropriate path and avoid the excessive congestion of a single link. Closer decouples the system into centralized route calculation and distributed route decision to guarantee its efficiency and stability in large-scale networks. In tested experiments, Closer achieves 2–7 times better flow completion time (FCT) than existing alternatives at 70% network load.