Analyzing the theoretical merits of Loxi load balancer for improving the efficiency of load balancing in 5G-edge IoT applications based on Kubernetes

Abstract

Load balancing, a critical aspect of cloud and cloud-based applications, is a major challenge that demands our attention. Due to the increasing dynamic workloads, load balancing becomes more important in the cloud. One of the hyperscale models that stands out for its ability to efficiently balance load by scaling the demands and allocating resources is the Loxi-Load-Balancer (LLB). This paper explores explicitly LLB's application in the context of 5G-Edge IoT applications based on Kubernetes. LLB's unique features, such as its open-source nature for cloud-native loads, its use of eBPF as the core engine to avoid adding additional software modules to configure the kernel, and its ability to change its services using the existing layers, set it apart from other load balancers. These features provide high security, observability, and networking. This paper delves into how LLB is used for load balancing in Kubernetes to increase speed and provide flexibility and customizable services. LLB automates all the internal and external administrations concerning monitoring, deployment, scaling, migration, routing, configuration, and resource allocation. This paper focused on developing an efficient resource allocation management system by load balancing using Loxi-Load-Balancer-extended Berkeley Packet Filter (LLB-eBPF). Detailed information about the LLB-eBPF-Kubernetes is given in this paper to help you understand the basics of LLB, eBPF, and Kubernetes.