North-South Load Balancing of Kubernetes Services with eBPF/XDP Martynas Pumputis (Isovalent) October 28, 2020 10.0.0.1 10.0.0.2 10.0.0.3 httpd httpd “httpd” service 10.0.0.1:30000 10.0.0.2:30000

individual containers getting started or destroyed as the application scales out / in to adapt to load changes and during rolling updates that are deployed as part of continuous delivery. This shift toward containers causes these approaches to struggle to scale side by side with the application as load balancing tables and access control lists carrying hundreds of thousands of rules that need to be updated with cloud network routers If you are already running routing daemons Load Balancing Cilium implements distributed load balancing for traffic between application containers and to external services and

individual containers getting started or destroyed as the application scales out / in to adapt to load changes and during rolling updates that are deployed as part of continuous delivery. This shift toward containers causes these approaches to struggle to scale side by side with the application as load balancing tables and access control lists carrying hundreds of thousands of rules that need to be updated conjunction with cloud network routers If you are already running routing daemons Load balancing Distributed load balancing for traffic between application containers and to external services. The loadbalancing

individual containers getting started or destroyed as the application scales out / in to adapt to load changes and during rolling updates that are deployed as part of continuous delivery. This shift toward containers causes these approaches to struggle to scale side by side with the application as load balancing tables and access control lists carrying hundreds of thousands of rules that need to be updated with cloud network routers If you are already running routing daemons Load Balancing Cilium implements distributed load balancing for traffic between application containers and to external services and

individual containers getting started or destroyed as the application scales out / in to adapt to load changes and during rolling updates that are deployed as part of continuous delivery. This shift toward containers causes these approaches to struggle to scale side by side with the application as load balancing tables and access control lists carrying hundreds of thousands of rules that need to be updated with cloud network routers If you are already running routing daemons Load Balancing Cilium implements distributed load balancing for traffic between application containers and to external services and

individual containers getting started or destroyed as the application scales out / in to adapt to load changes and during rolling updates that are deployed as part of continuous delivery. This shift toward containers causes these approaches to struggle to scale side by side with the application as load balancing tables and access control lists carrying hundreds of thousands of rules that need to be updated conjunction with cloud network routers If you are already running routing daemons Load balancing Distributed load balancing for traffic between application containers and to external services. The loadbalancing

individual containers getting started or destroyed as the application scales out / in to adapt to load changes and during rolling updates that are deployed as part of continuous delivery. This shift toward containers causes these approaches to struggle to scale side by side with the application as load balancing tables and access control lists carrying hundreds of thousands of rules that need to be updated conjunction with cloud network routers If you are already running routing daemons Load balancing Distributed load balancing for traffic between application containers and to external services. The loadbalancing

with individual containers ge�ng started or destroyed as the applica�on scales out / in to adapt to load changes and during rolling updates that are deployed as part of con�nuous delivery. This shi� toward of containers causes these approaches to struggle to scale side by side with the applica�on as load balancing tables and access control lists carrying hundreds of thousands of rules that need to be updated conjunc�on with cloud network routers If you are already running rou�ng daemons Load balancing Distributed load balancing for traffic between applica�on containers and to external services. The loadbalancing

helps improve the performance and reliability of NAS. It establishes multiple links for I/O load balancing and automatically switches I/Os to other available paths in case of a link failure, ensuring addresses, achieving functions such as multi-path link setup, fault recovery and failover, and load balancing. eNFS: Building an E2E High-Performance and Reliable Distributed File System for Production sysfs topology export Scheduling domain setup Topology-aware scheduling Migration and wakeup Load balancing Server Server 016 openEuler OS Technical White Paper Innovation Projects Application

10 05 11 06 12 Optimized process scheduling: The optimized load balancing algorithm decreases the overheads caused by load balancing between processes and improves application performance. Optimized CPU scheduling Memory management Cache QoS Network QoS Online preemption Offline kill SMT Load balancing Priority-based memory reclamation against OOM Dynamic watermark control policy Cache inference China by 2025, and centralized cloud computing will no longer be able to meet the necessary bandwidth load, network latency, and data management costs. In short, edge computing is vital to many industries