Skip to content
SRv6 TEK Innovations

Blog

SRv6 technology related blogs posted by SRv6 TEK Innovations Principal Network Architect Gyan Mishra to help facilitate discussion topics on SRv6 technology and how it is an industry wide network disrupter.

Log in to write a post or join the discussion.

316 posts

  1. Post 220 Restructured version · 🌐 Category 2: Network Fabric Revolution — The Deconstruction of Fixed Chassis

    The Pizza Box Ascendancy: Engineering High-Density 1RU/2RU Non-Blocking Switches

    💰 Silicon Efficiency: Processing Terabit-Scale SRv6 Pipelines on Merchant Silicon
  2. Post 219 Restructured version · 🌐 Category 2: Network Fabric Revolution — The Deconstruction of Fixed Chassis

    Monolithic Failure: The Financial and Operational Risks of Chassis Routers

    💰 Deconstructing the Giant: Replacing Single-Vendor Chassis with Resilient SRv6 Grids
  3. Post 218 Restructured version · 🌐 Category 1: The Compute Mutation — Beyond Blade Servers to DPU Enclaves

    DPUs in the AI Era: Orchestrating East-West Traffic for GPU Clusters

    💰 Unleashing AI Scale: Optimizing GPU Cluster Latency via DPU-Assisted SRv6
  4. Post 217 Restructured version · 🌐 Category 1: The Compute Mutation — Beyond Blade Servers to DPU Enclaves

    Orchestrating the DPU Control Plane: Canonical MAAS and OpenStack Ironic

    💰 Automated Provisioning: Orchestrating Dual-Plane Bare-Metal Nodes with SRv6 Profiles
  5. Post 216 Restructured version · 🌐 Category 1: The Compute Mutation — Beyond Blade Servers to DPU Enclaves

    Storage Disaggregation via DPU: Local NVMe Performance over NVMe-oF

    💰 Unified Fabric: Merging Storage Disaggregation and SRv6 Transport Networks
  6. Post 215 Restructured version · 🌐 Category 1: The Compute Mutation — Beyond Blade Servers to DPU Enclaves

    Physical Hardware Isolation: Enforcing Zero-Trust at the Server Edge

    💰 Un-Hackable Security: Hardening Edge Firewalls Using Isolated SRv6 Path Enclaves
  7. Post 214 Restructured version · 🌐 Category 1: The Compute Mutation — Beyond Blade Servers to DPU Enclaves

    Bare-Metal Deployment Flavor 2: Offloading Type-1 Hypervisors (KVM & ESXi)

    💰 Reclaiming Lost Power: Offloading Hypervisor SRv6 Encapsulation to Smart Silicon
  8. Post 213 Restructured version · 🌐 Category 1: The Compute Mutation — Beyond Blade Servers to DPU Enclaves

    Bare-Metal Deployment Flavor 1: True Native OS with DPU Multi-Tenancy

    💰 Multi-Tenant Agility: Injecting SRv6 Transport Segments directly onto Bare-Metal Hosts
  9. Post 212 Restructured version · 🌐 Category 1: The Compute Mutation — Beyond Blade Servers to DPU Enclaves

    Architecture Anatomy: The DPU as an Independent Mini-Server Pod

    💰 Silicon Evolution: Running Native SRv6 Pipelines Inside Isolated Server DPUs
  10. Post 211 Restructured version · 🌐 Category 1: The Compute Mutation — Beyond Blade Servers to DPU Enclaves

    The Death of the Midplane: Why the Blade Server Market Collapsed

    💰 The Compute Transformation: Moving from Rigid Blade Chassis to SRv6 Edge Nodes
  11. Post 247 · 🌐 Category 5: Host and Network Convergence — eBPF, DPDK, and Next-Gen SRv6

    Network and Server Convergence: Managing Server Clusters as One Routing Grid

    ⚡ The Technical Shift: Collapsing the traditional wall between systems administrators and network engineering teams by extending BGP and SRv6 down to individual DPU enclaves.
  12. Post 246 Original version · 🌐 Category 5: Host and Network Convergence — eBPF, DPDK, and Next-Gen SRv6

    The PINS Initiative: Integrating P4 with SONiC and SAI Control Planes

    ⚡ The Technical Shift: Exploring the P4 Integrated Network Stack (PINS) project, which models the standard SAI pipeline as a P4 program to allow remote P4Runtime orchestration.
  13. Post 245 Original version · 🌐 Category 5: Host and Network Convergence — eBPF, DPDK, and Next-Gen SRv6

    Next-Generation SRv6: Network-Wide Programmatic Policy Injection

    ⚡ The Technical Shift: Using Segment Routing over IPv6 (SRv6) to embed explicit service paths, traffic engineering policies, and compute instructions right into standard IPv6 headers.
  14. Post 244 Original version · 🌐 Category 5: Host and Network Convergence — eBPF, DPDK, and Next-Gen SRv6

    High-Performance User-Space Networking: The Role of DPDK

    ⚡ The Technical Shift: How the Data Plane Development Kit (DPDK) bypasses the standard Linux kernel network stack to talk directly to network hardware from user-space applications.
  15. Post 243 Original version · 🌐 Category 5: Host and Network Convergence — eBPF, DPDK, and Next-Gen SRv6

    The Host-Side Pipeline: eBPF vs. P4 for Internal Kernel Packet Processing

    ⚡ The Technical Shift: Comparing eBPF (running inside the Linux kernel) with P4 (running on hardware ASICs) for programmable network control.
  16. Post 242 Original version · 🌐 Category 4: Data Plane Engineering — P4, NPL, and Programmatic Pipelines

    The Hybrid Switch: Combining P4 Switching with Marvell OCTEON DPUs

    ⚡ The Technical Shift: Technical tear-down of hybrid switches (like Asterfusion) that couple an Intel Tofino or Broadcom switch chip with Marvell OCTEON processors on one internal bus.
  17. Post 241 Original version · 🌐 Category 4: Data Plane Engineering — P4, NPL, and Programmatic Pipelines

    Control Plane Interfacing: Comparing P4Runtime with Traditional SDK APIs

    ⚡ The Technical Shift: Architectural comparison of P4Runtime (a runtime control API) against traditional vendor Software Development Kit (SDK) binary interfaces.
  18. Post 238 Original version · 🌐 Category 4: Data Plane Engineering — P4, NPL, and Programmatic Pipelines

    Hardcoded Optimization: Using Scale Profiles and UFT Over Custom Coding

    ⚡ The Technical Shift: How fixed-function chips optimize resources via Unified Forwarding Tables (UFT) and Scale Profiles (e.g., L2-heavy vs. L3-heavy templates).
  19. Post 237 Original version · 🌐 Category 4: Data Plane Engineering — P4, NPL, and Programmatic Pipelines

    Broadcom's Counter-Strategy: Network Programming Language (NPL) Deep Dive

    ⚡ The Technical Shift: Architecture review of Broadcom's Network Programming Language (NPL) designed to optimize their programmable Trident chips.
  20. Post 236 Original version · 🌐 Category 4: Data Plane Engineering — P4, NPL, and Programmatic Pipelines

    The ASIC Oligopoly: How P4 Support Democratized Chip Architectures

    ⚡ The Technical Shift: Tracking how P4 moved from an Intel Tofino-exclusive feature to a language supported across Broadcom, Cisco Silicon One, Nokia FP, and Juniper families.