Networking Tech with Fexingo: Internet Infrastructure, Routing, and Network Engineering

Episode 42 of Networking Tech with Fexingo dives into Time-Sensitive Networking (TSN), the IEEE standard that guarantees microsecond-level latency for industrial and automotive applications. Lucas explains how TSN works with gating schedules and time-aware shaping, using a real-world example from a German automotive assembly line that cut cycle times by 15%. Luna questions whether TSN can scale beyond factory floors, and they explore its role in 5G fronthaul and autonomous vehicle platooning. The episode also covers the 802.1Qbv standard and how TSN differs from traditional best-effort Ethernet.

#TimeSensitiveNetworking #IndustrialEthernet #IEEE8021 #LowLatency #IndustrialAutomation #Automotive #AutonomousDriving #5GFronthaul #NetworkEngineering #TechPodcast #FexingoBusiness #BusinessPodcast #Technology #NetworkingTech #TSN #Ethernet #Manufacturing #Latency

Keep every episode free: buymeacoffee.com/fexingo

Episode 41 of Networking Tech with Fexingo explores why traditional network architectures buckle under AI training and inference workloads. Lucas and Luna break down the bottleneck: GPU clusters communicating via remote direct memory access (RDMA) over Converged Ethernet, and how a single packet loss can collapse training throughput by 50%. They examine the rise of co-designed networks for hyperscaler AI clusters—like Meta's Grand Teton and NVIDIA's DGX SuperPOD—and how InfiniBand is staging a comeback in data centers. The episode also covers how congestion control algorithms like DCQCN and new standards like Ultra Ethernet Consortium aim to fix the 'tail latency' problem that stalls large language model training. A must-hear for network engineers and AI infrastructure builders alike.

#AIWorkloads #NetworkArchitecture #RDMA #InfiniBand #UltraEthernet #DCQCN #MetaGrandTeton #NVIDIADGX #GPUNetworking #DataCenterNetworking #CongestionControl #TailLatency #LLMTraining #Hyperscaler #Technology #NetworkingTech #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Episode 40 of Networking Tech with Fexingo dives into EVPN-VXLAN, the technology modernizing data center interconnection. Lucas and Luna break down a real-world case: a major bank consolidating two legacy data centers into a single logical fabric. They explore how EVPN replaces spanning tree protocol for any-to-any connectivity, why VXLAN solves VLAN scalability limits for multi-tenant environments, and what the migration taught engineers about configuration complexity. The conversation covers practical examples—how the bank cut provisioning time from weeks to hours, why traffic engineering with BGP EVPN route types matters, and the surprising gotcha with ARP suppression. No jargon for jargon's sake; this is about understanding why EVPN-VXLAN is the de facto standard for modern data center fabrics. Listeners will walk away with a clear mental model of how control-plane-based overlays work and why they're essential for cloud-native architectures.

#EVPN #VXLAN #DataCenter #Networking #BGP #OverlayNetworking #DataCenterInterconnect #NetworkEngineering #SpineLeaf #Technology #FexingoBusiness #BusinessPodcast #Lucas #Luna #BankConsolidation #NetworkModernization #VLAN #Underlay

Keep every episode free: buymeacoffee.com/fexingo

Episode 39 dives into the mechanics of link-state routing protocols like OSPF and IS-IS when a fiber cut happens. Using the real-world example of a multi-vendor outage in Frankfurt in late 2025, Lucas and Luna walk through the convergence sequence: failure detection, LSA flooding, SPF recalculation, and FIB update. They explain why convergence times vary from milliseconds to seconds depending on technology choices like Bidirectional Forwarding Detection (BFD) and incremental SPF. The episode also covers the trade-off between fast convergence and network stability, including how timer back-off mechanisms prevent micro-loop storms. A concrete look at what happens inside routers when a trench gets dug in the wrong place.

#LinkStateRouting #OSPF #ISIS #BFD #SPF #Convergence #FiberCut #NetworkEngineering #RouteRedistribution #MicroLoop #FIB #LSA #FrankfurtOutage #Technology #FexingoBusiness #BusinessPodcast #InternetInfrastructure #RoutingProtocols

Keep every episode free: buymeacoffee.com/fexingo

Episode 38 of Networking Tech with Fexingo explores Zero Touch Provisioning (ZTP), the automation technology that lets network devices configure themselves without manual intervention. Lucas and Luna break down how ZTP works using DHCP and bootstrap servers, why it's essential for large-scale data center deployments, and how it reduces human error. They discuss real-world adoption by cloud providers and enterprises, compare ZTP to legacy manual provisioning, and touch on security considerations like authenticated bootstrapping. The episode also covers how ZTP fits into the broader network automation ecosystem alongside tools like Ansible and Cisco DNA Center. A must-listen for network engineers and anyone curious about how modern networks are built at scale.

#ZeroTouchProvisioning #NetworkAutomation #DHCP #BootstrapServer #DataCenter #Cisco #Ansible #CiscoDNACenter #NetworkEngineering #Technology #Automation #CloudNetworking #EnterpriseNetworking #ConfigurationManagement #NetworkDeployment #FexingoBusiness #BusinessPodcast #NetworkingTech

Keep every episode free: buymeacoffee.com/fexingo

Episode 37 of Networking Tech with Fexingo. Lucas and Luna explore network configuration drift—the silent accumulation of undocumented changes that leads to unexpected outages. They walk through a real 2025 incident where a single uncommitted 'no shut' command on a core router caused a 47-minute outage for a major European bank. They explain what drift is, how it happens (manual patches, emergency changes, OS upgrades), and why traditional auditing tools catch only 60% of discrepancies. They also discuss emerging solutions: intent-based networking, automated backup validation, and immutable infrastructure. The episode closes with a practical takeaway: any network engineer can start a simple 'golden config' diff process this week without buying new software.

#NetworkConfigurationDrift #NetworkOutages #RouterConfig #IntentBasedNetworking #GoldenConfig #NetworkAutomation #NetworkEngineering #NetworkReliability #TechIncident #EuropeanBank #ConfigurationManagement #ImmutableInfrastructure #NetworkAudit #CLI #Ansible #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Ethernet is the quiet backbone of the internet, but it's not standing still. In this episode, Lucas and Luna explore how Ethernet is evolving beyond the data center to meet the demands of 5G, industrial IoT, and autonomous systems. They focus on the IEEE 802.3cz standard for 800 Gigabit Ethernet, the rise of Time-Sensitive Networking (TSN) for deterministic latency, and how companies like Cisco and Arista are pushing Ethernet into new realms like in-vehicle networking and factory floors. Luna challenges whether Ethernet can really replace dedicated fieldbus systems in manufacturing, while Lucas argues that its ubiquity and cost advantages make it inevitable. The conversation lands on a specific example: BMW's use of TSN over Ethernet for real-time control in its assembly lines. If you think Ethernet is just a cable in your office wall, this episode will change your mind.

#EthernetEvolution #8023cz #800GigabitEthernet #TimeSensitiveNetworking #TSN #IndustrialIoT #AutonomousSystems #BMW #Cisco #Arista #NetworkEngineering #Technology #FexingoBusiness #BusinessPodcast #InternetInfrastructure #Routing #DeterministicNetworking #IEEE

Keep every episode free: buymeacoffee.com/fexingo

Episode 35 of Networking Tech with Fexingo dives into VXLAN (Virtual Extensible LAN) — the overlay protocol that lets network engineers stretch Layer 2 networks across distant data centers. Lucas and Luna unpack how VXLAN encapsulates Ethernet frames in UDP packets, enabling virtual machine mobility and multi-tenant isolation without rewriting physical infrastructure. They walk through a real scenario: a cloud provider migrating a live database workload between data centers in Ashburn and Dallas with zero downtime. The conversation covers VXLAN's 24-bit segment ID (16 million virtual networks vs. VLAN's 4096), the role of VTEPs (VXLAN Tunnel Endpoints), and why hardware offload matters for line-rate performance. They also touch on trade-offs — increased MTU overhead, multicast dependency in early implementations, and how EVPN (Ethernet VPN) now solves control-plane scaling. No marketing fluff, just clear engineering insight on one of the most important data-center networking technologies of the last decade.

#VXLAN #NetworkVirtualization #DataCenterNetworking #OverlayNetworks #VTEP #EVPN #Layer2Extension #CloudNetworking #Cisco #VMwareNSX #NetworkEncapsulation #UDP #MultiTenancy #VLAN #Technology #NetworkingTech #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna dive into segment routing, a modern traffic engineering approach that simplifies how networks steer packets. They explain how segment routing labels replace complex per-flow state, compare it to MPLS, and walk through a real-world case where a large cloud provider cut bandwidth costs by 30 percent. The episode covers the role of the segment routing header in IPv6, how it improves network utilization, and why engineers say it makes troubleshooting easier. They also touch on the inevitable trade-offs: hardware upgrades and a learning curve for teams used to traditional routing protocols. A concrete look at a technology quietly reshaping backbone and data center networks.

#SegmentRouting #SRv6 #TrafficEngineering #NetworkRouting #MPLS #IPv6 #SDN #NetworkOptimization #BackboneNetworks #DataCenter #TechExplainer #NetworkEngineering #InternetInfrastructure #Technology #FexingoBusiness #BusinessPodcast #NetworkingTech #PodcastEpisode

Keep every episode free: buymeacoffee.com/fexingo

Episode 33 of Networking Tech with Fexingo dives into Zettastructure, a radical network architecture that treats the entire data center fabric as a single logical switch. Lucas and Luna unpack how this approach, pioneered by startups like DriveNets and Arrcus, disaggregates hardware from software and uses centralized control planes to eliminate spine-leaf complexity. They walk through a real example: a 100,000-server cluster where traditional routing would require dozens of protocol sessions, but Zettastructure reduces it to just one. With references to white-box switching and the Open Config initiative, this episode explains why hyperscalers like Google and Meta are quietly adopting elements of this design. By June 2026, the approach is moving from edge labs to mainstream enterprise. Expect concrete numbers on cost savings (roughly 40% lower capital expenditure) and latency improvements (sub-10 microsecond fabric-wide). A must for network engineers evaluating next-gen data center designs.

#Zettastructure #DataCenterNetwork #Routing #DriveNets #Arrcus #WhiteBoxSwitching #OpenConfig #NetworkArchitecture #SpineLeaf #Hyperscaler #Google #Meta #CentralizedControlPlane #NetworkDisaggregation #Technology #FexingoBusiness #BusinessPodcast #InternetInfrastructure

Keep every episode free: buymeacoffee.com/fexingo