TopoGen - Network Topology Generator

Phase 22/25

Concept

Generating realistic, validated network topologies for testing often requires custom, brittle scripts. The Topology Generator consolidates topology generation logic into a Rust core, producing structured data center and WAN graphs with consistent YAML output for the simulation and visualization ecosystem.

Quick Facts


Status	Phase 22/25
Language	N/A
Started	2026

What This Is

A Rust library with Python bindings for generating realistic network topologies. Consolidates scattered topology generation logic from AutoNetKit, simulation tools, and visualization tools into one well-tested implementation.

Key Features

Topology Types:
- Data center: Fat-tree, leaf-spine
- WAN/Backbone: Ring, mesh, hierarchical WAN, POP (Points of Presence)
- Random graphs: Barabási-Albert, Watts-Strogatz, Erdős-Rényi
Realistic Interface Specifications: Vendor-specific naming conventions (Cisco, Arista, Juniper)
Traffic Matrix Generation: Gravity-model traffic matrices with temporal dynamics (diurnal/weekly patterns)
Distance-Aware Latency: Geographic placement drives realistic link latencies
Seed-Deterministic: Reproducible topology generation from a seed value
Output Converters: YAML, ContainerLab topology files, AutoNetKit GraphML
Three Interfaces: CLI, Python API, and config-driven YAML/TOML input
Validation: Structural and design-pattern compliance checks

Example: Data Center Leaf-Spine

Configuration file (leaf-spine-lab.yaml):

name: dc-lab-leaf-spine-2s-4l-100g
seed: 42

type: leaf-spine
spines: 2
leaves: 4
full_mesh: true
spine_bandwidth_gbps: 100.0

Generate topology:

$ topogen generate leaf-spine-lab.yaml --output topology.yaml

Generated topology: dc-lab-leaf-spine-2s-4l-100g
  Nodes: 6 (2 spines, 4 leaves)
  Links: 8 (full mesh spine-leaf connectivity)
  Total bandwidth: 800 Gbps
  Oversubscription: 2:1 (standard)

Output written to: topology.yaml

Generated output (excerpt):

name: dc-lab-leaf-spine-2s-4l-100g
topology_type: leaf-spine

nodes:
  - name: spine-1
    type: spine
    tier: 1
  - name: spine-2
    type: spine
    tier: 1
  - name: leaf-1
    type: leaf
    tier: 2
  - name: leaf-2
    type: leaf
    tier: 2
  - name: leaf-3
    type: leaf
    tier: 2
  - name: leaf-4
    type: leaf
    tier: 2

links:
  - src: leaf-1
    dst: spine-1
    bandwidth_gbps: 100.0
    latency_ms: 0.1
  - src: leaf-1
    dst: spine-2
    bandwidth_gbps: 100.0
    latency_ms: 0.1
  # ... (full mesh: 4 leaves × 2 spines = 8 links)

Python API

from topogen import DataCenter

topo = DataCenter.spine_leaf(
    spines=2,
    leaves=4,
    full_mesh=True,
    spine_bandwidth_gbps=100.0
)

topo.export_yaml("topology.yaml")
topo.export_json("topology.json")

# TopoGen - Network Topology Generator
from ank_pydantic import Topology
ank_topo = topo.to_ank_pydantic()

Milestones

v1.0-alpha Foundation (Shipped Jan 28, 2026) Core topology generators, validation framework, and output format.

Data center generators (fat-tree, leaf-spine) and WAN generators (ring, mesh, hierarchical)
Random graph generators (Barabási-Albert, Watts-Strogatz, Erdős-Rényi)
Structural and design-pattern validation
YAML output format

v0.9 User Interfaces (Shipped Feb 5, 2026) CLI, Python API, config-driven generation, and documentation.

Production CLI with shell completions and JSON diagnostics
Python bindings via PyO3 with full API parity
Config-driven generation from YAML/TOML files
Documentation site with algorithm references and example gallery

v0.10 Gap Closure (Shipped Feb 11, 2026) Output converters, traffic patterns, and POP design patterns.

ContainerLab topology file export
AutoNetKit GraphML converter
Gravity-model traffic matrices with diurnal/weekly temporal dynamics
POP design patterns — access, metro, regional, and backbone tiers with geographic placement and redundancy models

v1.0 Production Features (In Progress — Phases 22-24)

Geographic placement infrastructure — City database lookup, H3 grid placement, distance-based latency calculation, region/country filtering
Fiber map integration — GeoJSON import/export, estimated vs actual fiber route distinction, terrain plausibility checks
Multi-layer topology generation — Explicit underlay/overlay layer dependencies, overlay-to-underlay path mapping, cross-layer capacity validation
Eyeball network generator — Access-aggregation-core hierarchy for ISP topologies, inbound-heavy traffic ratios, peering and transit policy configuration

Tech Stack

Rust core, PyO3 for Python bindings, Maturin build system

← Back to Network Automation