Unpacking “Edge” In Verizon 5GEdge
summary
Verizon describes a deployed cloud‑native edge with thousands of far-edge sites (20x20 closets and single 2U servers up to pooled CDN sites), running heterogeneous CPUs, GPUs and FPGA/ASIC accelerators, PTP for nanosecond sync, and an open‑source CNCF-inspired stack. Design goals include platform overhead under 10%, local control planes loosely coupled to a central controller that still needs scaling to manage 10,000s of sites, zero touch provisioning, and minute-scale resiliency and parallel upgrades — already running real traffic.
AI-written summary · about
chapters (12)
- 0:00Intro
- 1:05The Promise of 5G
- 2:42Achieving Openness Through Virtualization
- 3:52Verizon Network Cloud Topology
- 6:04Use Cases Supported By Edge Infrastructure
- 7:06Edge Technology Tenets
- 9:51Maximize Compute and Minimize Platform Overhead
- 12:30Common Platform For Applications
- 13:57Deployment @ Scale
- 14:40Distributed Edge Platform Architecture
- 16:36Resiliency and Recovery
- 17:56Key Takeaways
related talks
44mHow 5G Edge is being used to solve customer problems
07mScalable Edge Data Centers: Current Challenges and How to …
44mEdge Computing ... What's That All About?
42mThe Challenges and Solutions of Open Edge Infrastructures
36mOpenEdge, the General Framework of Edge Computing
41mCloud Computing: The Edge talks - long road into the Cloud
36mEdge Computing Is Hot. Find Out the Business Value From Three Experts
35mPCEI at the EDGE