Files
2026-06-13 19:50:42 +05:30

6.6 KiB
Raw Permalink Blame History

Architecture Analyzer

ComfyFlow Insight

Technical Documentation


Overview

The Architecture Analyzer is one of the core engines inside ComfyFlow Insight.

Its responsibility is to inspect workflow structure and generate architectural insights that help users understand how a workflow is organized.

Unlike workflow execution analysis, Architecture Analysis focuses on structural design.

It answers questions such as:

  • How large is the workflow?
  • How deep is the execution chain?
  • Where does execution begin?
  • Where does execution end?
  • How many branches exist?
  • How complex is the workflow architecture?

Why Architecture Analysis Matters

As workflows grow larger, understanding the overall structure becomes increasingly difficult.

A workflow containing:

150 Nodes
220 Links
8 Branches

cannot be understood by visually inspecting the canvas alone.

Architecture Analysis provides an engineering-level view of workflow design.


Objectives

The Architecture Analyzer performs the following tasks:

  • Detect Entry Nodes
  • Detect Exit Nodes
  • Calculate Workflow Depth
  • Calculate Longest Path
  • Detect Branching Structure
  • Estimate Architecture Complexity
  • Generate Architecture Summary

Workflow Architecture Model

The analyzer treats a workflow as a directed graph.

Example:

LoadImage
    ↓
ImageScale
    ↓
KSampler
    ↓
SaveImage

Graph Representation:

Node A → Node B → Node C → Node D

Each node becomes a graph vertex.

Each link becomes a graph edge.


Entry Nodes

Definition

Entry Nodes are nodes that do not depend on any other nodes.

These nodes represent workflow starting points.


Example

LoadImage
CheckpointLoader
TextPrompt

These nodes receive no workflow input.


Why Entry Nodes Matter

Entry Nodes reveal:

  • Workflow starting locations
  • Independent execution paths
  • Data sources

Large numbers of Entry Nodes may indicate:

  • Multiple workflow sections
  • Parallel processing pipelines
  • Workflow fragmentation

Exit Nodes

Definition

Exit Nodes are nodes that do not feed into any other node.

These nodes represent workflow endpoints.


Example

SaveImage
VideoCombine
OutputNode

Why Exit Nodes Matter

Exit Nodes identify:

  • Workflow outputs
  • Final processing stages
  • Delivery points

Unexpected Exit Nodes often indicate:

  • Incomplete workflows
  • Missing connections
  • Broken output chains

Workflow Depth

Definition

Workflow Depth measures the maximum execution distance from an Entry Node to an Exit Node.


Example

LoadImage
    ↓
ImageScale
    ↓
KSampler
    ↓
SaveImage

Depth:

4

Interpretation

1 – 10      Simple
11 – 25     Medium
26 – 50     Advanced
50+         Enterprise Scale

Why Depth Matters

Deeper workflows:

  • Require more maintenance
  • Increase debugging difficulty
  • Create hidden dependencies
  • Increase execution complexity

Longest Path

Definition

Longest Path measures the largest execution chain within the workflow.


Example

LoadImage
    ↓
Resize
    ↓
ControlNet
    ↓
KSampler
    ↓
Upscaler
    ↓
SaveImage

Longest Path:

6 Nodes

Benefits

Longest Path identifies:

  • Critical execution routes
  • Workflow bottlenecks
  • Processing complexity

Branch Detection

Definition

A branch occurs when one node feeds multiple downstream nodes.


Example

ImageScale
    ├── SaveImage
    ├── PreviewImage
    └── VideoCombine

Branch Count:

3

Branch Importance

Branches increase:

  • Execution complexity
  • Workflow size
  • Maintenance effort

Branches also improve:

  • Flexibility
  • Reusability
  • Multi-output workflows

Architecture Complexity

Architecture Complexity is derived from:

Node Count

Link Count

Workflow Depth

Longest Path

Branch Count

Example

Workflow:

Nodes      = 120
Links      = 180
Depth      = 28
Branches   = 7

Result:

Architecture Complexity = High

Architecture Categories

Simple

Characteristics:

Few Nodes

Few Connections

Minimal Branching

Typical Size:

1 – 25 Nodes

Medium

Characteristics:

Multiple Processing Stages

Moderate Branching

Manageable Dependencies

Typical Size:

25 – 75 Nodes

Advanced

Characteristics:

Complex Processing

Multiple Branches

Large Dependency Chains

Typical Size:

75 – 200 Nodes

Enterprise

Characteristics:

Large Pipelines

Extensive Branching

High Dependency Density

Team Collaboration

Typical Size:

200+ Nodes

Architecture Summary

The Architecture Analyzer generates a high-level architectural summary.

Example:

Workflow Type:
Advanced

Entry Nodes:
3

Exit Nodes:
2

Workflow Depth:
28

Longest Path:
31

Branch Count:
7

Architecture Complexity:
High

Common Findings

Too Many Entry Nodes

Possible Causes:

  • Workflow fragmentation
  • Multiple independent systems

Recommended Action:

  • Review workflow organization

Too Many Exit Nodes

Possible Causes:

  • Multiple outputs
  • Incomplete workflow design

Recommended Action:

  • Verify intended outputs

Excessive Depth

Possible Causes:

  • Overengineering
  • Unnecessary processing stages

Recommended Action:

  • Simplify workflow where possible

Excessive Branching

Possible Causes:

  • Workflow duplication
  • Parallel processing

Recommended Action:

  • Review branch necessity

Best Practices

Healthy Architecture

Characteristics:

Clear Entry Nodes

Clear Exit Nodes

Controlled Branching

Reasonable Depth

Organized Dependencies

Relationship With Other Engines

Architecture Analyzer works closely with:

Workflow Statistics

Branch Analyzer

Dependency Graph

Workflow Advisor

Together they provide a complete engineering view of workflow design.


Summary

The Architecture Analyzer is responsible for understanding workflow structure.

It converts raw workflow graphs into meaningful architectural insights by identifying:

  • Entry Nodes
  • Exit Nodes
  • Workflow Depth
  • Longest Path
  • Branching Structure
  • Architecture Complexity

These metrics form the foundation for all higher-level analysis performed by ComfyFlow Insight.


Next Document

Continue with:

docs/workflow-statistics.md

This document explains workflow metrics, scoring systems, density calculations, complexity measurements, and workflow scale classification.