Kube-OVN v1.16.0 introduces major enhancements such as BGP/EVPN‑enabled VPC egress, a tiered SecurityGroup with expanded priority range, per‑NIC DHCP control, multi‑network NetworkPolicy annotations, full‑NIC hot migration for KubeVirt, static IP/MAC per interface, and numerous reliability, performance, and Helm chart upgrades.
iQIYI replaced its limited private‑cloud DNS by a hybrid‑cloud Anycast solution that uses an Enterprise Router with BGP‑advertised Anycast IPs and GRE tunnels to integrate public‑cloud DNS nodes, delivering unified, elastic domain resolution, rapid failover, and centralized management across IDC and cloud environments.
This article explains the core principles, advantages, suitable scenarios, and step‑by‑step configuration of Calico's BGP mode, showing how it delivers high‑performance, scalable, and secure networking for large Kubernetes deployments.
This article introduces MetalLB, an open‑source load‑balancing solution for self‑managed Kubernetes clusters, explains its deployment requirements, operating principles, Layer 2 and BGP modes, and provides step‑by‑step installation and verification instructions using Kubernetes manifests, Helm, and Kustomize.
This article explains Kubernetes' container networking fundamentals, covering the basic network requirements, IPAM implementation with node CIDR allocation, Linux veth and bridge mechanisms, overlay solutions like VXLAN, and the role of BGP for cross‑cluster pod reachability.
The article examines the history, current relevance, vendor support, and practical testing of IPv4 Class E address space, evaluating its potential as a local unicast pool amid IPv4 exhaustion and the challenges of deploying it with routing protocols like OSPF and BGP.
This article explores the history, technical feasibility, vendor support, and real‑world testing of the unused IPv4 Class E address block (240.0.0.0/4), concluding that while it can serve large private networks, widespread adoption is impractical and IPv6 remains the preferred solution.
This report details a network operations incident where a BGP routing change caused an EBGP neighbor to go idle, outlines the step‑by‑step troubleshooting, analysis of the root cause, and the implemented solution involving a new L3 node and redundant EBGP peers.
This article introduces MetalLB, explains its deployment requirements, describes its Layer2 and BGP operation modes, provides step‑by‑step installation and configuration instructions for Kubernetes clusters, and demonstrates verification of the load‑balancing functionality.
This article explains the fundamentals of EVPN, how it extends MP‑BGP to provide a control plane for VXLAN, details the five EVPN route types, and shows how MAC learning, head‑end replication, and ARP suppression work in a distributed VXLAN network.
This article explains the fundamentals of Ethernet VPN (EVPN), its MP‑BGP‑based routing types, and how EVPN serves as the control plane for VXLAN, covering route formats, MAC learning, head‑end replication, host and subnet route distribution, and ARP broadcast suppression techniques.
This guide walks through setting up a simulated Kubernetes cluster with Cilium, configuring BGP using FRR on a router node and Bird on worker nodes, and enabling Cilium's built‑in BGP speaker to achieve cross‑node pod communication and LoadBalancer IP advertisement.
This article provides a detailed overview of Calico's essential and optional components, explaining how each part—such as the API server, Felix, BIRD, confd, Dikastes, CNI plugin, datastore plugin, IPAM plugin, kube‑controllers, Typha, and calicoctl—contributes to networking and network policy enforcement in Kubernetes clusters.
This article provides a comprehensive overview of dynamic routing protocols, detailing their basic operation, classification, common implementations such as RIP, OSPF, and BGP, performance metrics, BGP message structures, path attributes, route selection rules, and advanced concepts like route reflectors, multiprotocol extensions, and BGP‑LS for topology extraction.
This article explains how to build a reliable, high‑performance load‑balancing platform by analyzing basic reliability requirements, introducing multi‑layer (DNS, L4, L7) balancing, comparing Direct‑Route, Tunnel and FULLNAT modes, and describing the VGW architecture, health‑check, fault isolation, redundancy and DPDK‑based performance optimizations.
This article examines three real‑world routing loop incidents caused by static‑route misconfiguration, missing passive‑interface settings in OSPF, and improper bidirectional redistribution, and provides detailed analysis and practical recommendations to prevent such loops in enterprise network operations.
The article details how a multi‑layer load‑balancing platform (VGW) was designed—combining 7‑layer Nginx, 4‑layer LVS with FULLNAT, and 3‑layer network devices—to achieve business reliability, fault isolation via BGP‑announced VIPs, and high‑throughput performance using DPDK, while providing redundancy at server, link, and cluster levels.
This guide explains how to install the kube-ovn-speaker component, configure BGP parameters, modify subnet settings, add or remove annotations, and use advanced BGP options to publish Pod and Subnet routes externally in a Kubernetes cluster using Kube‑OVN.
This guide explains why a Calico pod shows "BIRD is not ready: BGP not established" in a Kubernetes cluster, identifies the missing node interface detection as the root cause, and provides step‑by‑step instructions to adjust the IP autodetection method and restore pod health.
To cut the hundreds‑of‑thousands‑yuan monthly bill of a commercial HTTPDNS service, the team built a multi‑region, self‑hosted HTTPDNS platform, estimated to slash costs by up to 90%, then resolved unexpected TLS bandwidth waste by improving connection reuse, ultimately achieving over 80% savings and planning a hybrid‑cloud deployment.
This article explains Calico's IPIP networking mode in Kubernetes, covering its architecture, core components, routing behavior, command‑line examples, and packet‑capture analysis to illustrate how pods communicate across nodes.
This article explains how EVPN serves as the control plane for VXLAN, detailing its architecture, BGP route types, advantages like reduced flooding and simplified configuration, and the mechanisms for tunnel establishment, MAC learning, routing, and ARP suppression in modern data‑center networks.
This article details the step‑by‑step evolution of a multi‑cloud network at Zuoyebang, covering three construction phases, quality‑improvement measures such as fault prevention and rapid recovery, and ongoing operational governance that together deliver a flexible, high‑availability cloud infrastructure.
This article explains the fundamentals of Ethernet Virtual Private Network (EVPN), covering its basic concepts, topology options, control‑plane mechanisms, forwarding processes for unicast and multicast traffic, and the key advantages such as load balancing, reduced mesh overhead, faster convergence, and minimized ARP flooding.
This article explains why exposing services on a bare‑metal Kubernetes cluster requires a custom load‑balancer solution, introduces MetalLB, details its deployment requirements, explains its Layer 2 and BGP modes, provides step‑by‑step installation commands, and shows how to verify the setup.
Bilibili's network team designed and evolved its data‑center fabric from a stacked Layer‑2 V1.0 architecture through an M‑LAG EBGP‑based V2.0 design to a uniform box‑type V3.0 deployment, achieving greater stability, scalability, cost efficiency, and operational simplicity via extensive ARP, LACP, DHCP, hash, and BGP optimizations.
This guide explains how to configure Calico networking in Kubernetes clusters, covering the supported modes (IPIP, BGP, cross‑subnet), route reflector setup, MTU tuning, AS number configuration, IP pool management, SNAT disabling, and assigning fixed pod IPs.
The article explains how Facebook engineered a scalable, automated public peering management system—replacing manual, error‑prone processes with a workflow that uses a dedicated peering portal, PeeringDB OAuth authentication, and automated BGP session provisioning to improve reliability and save hours each week.
Facing the limitation of LoadBalancer services on bare‑metal Kubernetes, this guide walks through the background, MetalLB overview, and step‑by‑step deployment using BGP, including environment setup, configuration files, and verification of external IP routing.
This article explains how Kubernetes ensures container-to-container communication using network namespaces, veth pairs, bridges like docker0, and advanced CNI plugins such as Flannel and Calico, detailing their underlying Linux networking components, routing mechanisms, and deployment considerations for both intra‑host and inter‑host scenarios.
This article explains Kubernetes networking fundamentals, including intra‑pod communication, pod‑to‑pod traffic, service IP routing, and compares CNI plugins like Flannel and Calico with details on IPIP and BGP optimizations.
This article details the background, design rationale, network topology changes, and step‑by‑step procedures—including VPC configuration, BGP setup, and port‑channel adjustments—used to upgrade Qunar's data‑center network for Kubernetes deployments, with practical code examples and operational tips.
This article explains how containers communicate within a pod, between pods, and with services in Kubernetes, and compares Calico's IPIP and BGP networking modes, detailing their traffic handling, suitable scenarios, efficiency, and configuration steps.
This comprehensive guide explains the purpose, types, metrics, and convergence behavior of major routing protocols—including static, dynamic, distance‑vector, and link‑state protocols such as OSPF, EIGRP, RIP, IS‑IS, and BGP—while detailing their selection algorithms, administrative distances, and key configuration concepts.
Cilium 1.11 introduces a beta Service Mesh, Kubernetes Ingress support, OpenTelemetry integration, topology‑aware load balancing, BGP pod‑CIDR announcements, managed IPv4/IPv6 neighbor discovery, XDP multi‑device acceleration, graceful termination, scalable ID spaces, endpoint slices and several feature enhancements and deprecations.
This guide walks through setting up a cross‑subnet Kubernetes cluster with Calico in pure BGP mode by simulating routers and networks using ENSP and a workstation, covering architecture design, ENSP routing configuration, node integration, Calico BGP settings, and verification of pod communication.
OpenELB, the CNCF‑sandboxed load‑balancer plugin from KubeSphere, offers BGP and Layer‑2 based traffic distribution, CRD‑driven IP pool management, and seamless integration with Kubernetes, K3s and edge environments, positioning itself as a lightweight, community‑driven alternative to MetalLB for on‑premise deployments.
This article introduces the Border Gateway Protocol (BGP), explains its role as the core decentralized routing protocol of the Internet, compares it with IGP and EGP, describes why and when BGP is needed in cloud environments, and details its TCP‑based operation, message types, and path‑vector mechanics.
Facebook experienced a six‑hour worldwide outage that knocked out its main site and services like Instagram, WhatsApp, Messenger and Oculus, and engineers later traced the incident to a misconfigured backbone router that broke BGP routing and DNS resolution, sparking conspiracy rumors about data leaks.
iQIYI’s SDN‑based full‑network automatic traffic‑scheduling system dynamically balances inter‑ and intra‑province traffic using BGP and policy routing, integrates monitoring, flow collection, DFS backup‑path calculation, and real‑time Kafka/Flink processing, cutting fault‑handling time to minutes and boosting link availability to 99.9999 % while preparing for programmable‑switch and SR‑based extensions.
This comprehensive guide explains when to employ multiple routing protocols, compares distance‑vector and link‑state routing, details access‑list types and creation commands, outlines route redistribution steps and commands, clarifies administrative distance, BGP fundamentals, metric concepts, and provides hardware compatibility notes for Cisco routers and switches.
This article explains how to build a high‑performance, 5K‑node hybrid‑cloud container network by reviewing Kubernetes pod and service networking, comparing iptables, IPVS and eBPF implementations, and describing large‑scale overlay, BGP, IPVlan, and NodeLocal DNS solutions.
This article details Qunar's adoption of Calico as a CNI solution, covering its three‑layer architecture, core components, large‑scale deployment practices, IPAM behavior, pod‑to‑pod and pod‑to‑external traffic flows, encountered issues, and the overall benefits for a production Kubernetes environment.
This article explains the fundamentals of Kubernetes container networking, covering the network stack, bridge, veth pairs, host networking, ARP resolution, and then dives into cross‑host communication methods, comparing Overlay solutions such as Flannel and Calico with Underlay approaches like BGP, illustrated with concrete commands and diagrams.
This guide explains how kube-proxy in iptables mode, together with Calico BGP routing, translates service IPs to pod IPs, creates custom iptables chains, and routes traffic from external or node ports to the correct pod across a Kubernetes cluster.
This article explains the fundamental principles of Kubernetes container networking, how Docker’s default bridge and veth pairs connect pods on the same host, and compares various CNI plugins and routing models for cross‑host communication, including Flannel and Calico.
This article describes Ctrip's adoption of Cilium with BGP to build a high‑performance, eBPF‑based cloud‑native network, covering the evolution from Neutron+OVS, the BGP peering model, traffic forwarding paths, L4 load‑balancing, security policy implementation, and multi‑cluster integration using ClusterMesh.
This article provides a comprehensive guide to Calico as a Kubernetes CNI, covering its BGP‑based three‑layer architecture, deployment steps, configuration of etcd certificates, management tools, BGP route‑reflector mode, IPIP tunneling, and network policy implementation for cloud‑native clusters.
This article examines Calico’s networking architecture, compares its IPIP and BGP modes, and presents detailed performance measurements—including ping latency, bandwidth via iperf, and HTTP load testing—across physical machines and Kubernetes pods, highlighting the impact of each mode on throughput and latency.
The article answers a technical question about using Calico as the network mode in container cloud networks, explaining that Calico relies on three‑layer routing with BGP, requiring hardware BGP support and sufficient processing capacity to host the full‑mesh routing for all nodes.
This article explains the principles of LVS load balancing, its benefits, the Hulk LVS cluster architecture, request hashing mechanisms, system‑level high availability, and the HULK LVS forwarding panel, providing a comprehensive guide for building resilient high‑traffic web services.
This article examines the evolution of data‑center server access networks, explains why traditional large Layer‑2 designs are problematic at scale, and presents a pure three‑tier underlay solution that uses host routing, ECMP, and ARP proxy to achieve seamless KVM communication without overlay overhead.
iQIYI’s SD‑WAN‑based Network Acceleration Platform builds an overlay over the existing underlay, separating business control, routing, and data‑forwarding layers to provide automated, policy‑driven acceleration for physical hosts, virtual machines and other services, while offering flexible inter‑node routing, unified management, and fault‑tolerant, QoS‑aware path selection.
On March 23, a construction accident severed a fiber optic cable in Shanghai, causing widespread internet disruptions, but Tencent Cloud’s intelligent traffic scheduling system and four‑fiber‑three‑router high‑redundancy architecture automatically rerouted traffic, restoring services within two minutes and demonstrating robust cloud network resilience.
Using a vivid Journey to the West allegory, this article walks through cloud networking fundamentals—from regions, availability zones, and VPCs to BGP routing, DNS resolution, TCP handshakes, TLS encryption, load balancing, and micro‑service communication—illustrating each concept with clear diagrams and analogies.
From Hearthstone’s dual‑database crash to Uber’s massive data breach, this 2017 Linux operations roundup chronicles twelve critical incidents—highlighting backup failures, Docker rebranding, ransomware, BGP hijacking, and more—offering key lessons for sysadmins and DevOps professionals.
The article explains how Alibaba’s Edge Network, using the NetO intelligent traffic scheduling system, combines real‑time internet quality monitoring, SDN‑enhanced BGP routing, and extensive peering to optimize both inbound and outbound paths for millions of global Double 11 shoppers, ensuring a seamless user experience.
This article provides a detailed overview of China Telecom’s extensive IDC infrastructure, explaining the hierarchy of its backbone and metropolitan networks, the roles of ChinaNet and CN2, BGP routing practices, inter‑carrier connections, and practical traffic‑optimization techniques used by both major and smaller operators.
The 2015 SACC conference highlighted numerous hidden challenges in building Internet+ network infrastructure—including cloud host, DNS, CDN, IP database, and BGP issues—providing practical insights for architects and IT professionals to avoid costly mistakes.
This article explores why robust network infrastructure is essential for cloud platforms, examines China’s telecom landscape, outlines inter‑carrier connectivity solutions such as multi‑line BGP, and provides a practical evaluation model to help businesses select the optimal network and data‑center for their cloud workloads.
