Synthesizing Plausible Infrastructure Configurations for Evaluating Edge Computing
summary
A Python tool called edge topology synthesizer generates parameterized infrastructure graphs (node types like Raspberry Pi 3 and Intel NUC, link bandwidths, hierarchical cells) and exports to NetworkX; it synthesizes urban scenarios from real data (about 200 Array of Things nodes in Chicago), fits node-density as log‑normal, and can instantiate municipal cloudlets (example: 10 servers, 2 racks connected by fiber) with uplink/downlink profiles. The tool feeds a custom network simulator and already produces usable topologies and latency graphs for evaluation.
AI-written summary · about
chapters (9)
- 0:00Intro
- 1:40Edge systems evaluation tools
- 3:22City-Scale Edge Computing Fabric
- 4:10Edge Topology Synthesizer
- 4:20Conceptual model
- 6:27Parameterized cell synthesis
- 7:37GeoCell Density
- 8:28Programmatic creation of topologies
- 9:50Interconnecting topologies
related talks
39mThorny path to the Large Scale Graph Processing
33mFeeding the Cloud: Designing Data-Driven Robotics Systems
50mBig Data on Smallish Machines
24mStreaming analytics on edge with Kuiper and KubeEdge
29mTed Willke, Senior Principal Engineer & GM, Datacenter Group, Intel
27mData Mesh for the Cognitive City
07mData Gravity and Kubernetes: Managing Large-Scale Data Ingest With Minimal Latency
27mData @scale Seattle - Yucheng Low