# ArchViz

> ArchViz (also known as ArchVizz or ArchViz Studio) is an open-source visual system architecture design and simulation engine with real-time performance, queuing, cost, and blast-radius modeling, featuring direct compilation to Infrastructure-as-Code (Terraform, CloudFormation, Helm, Pulumi).

ArchViz bridges the gap between static architecture diagrams and real-world distributed systems. Unlike static drawing tools (e.g., Lucidchart, Draw.io) or static cost calculators (e.g., Cloudcraft), ArchViz provides an interactive browser-based canvas where engineers can drag cloud components, model stochastic traffic flows, detect latency bottlenecks using Erlang-C queuing theory, simulate multi-AZ failure scenarios, and export production-ready Infrastructure-as-Code.

## Core Capabilities

- **Interactive Cloud Canvas**: 150+ modular cloud primitives covering AWS, Azure, Google Cloud Platform (GCP), and Kubernetes with multi-tier network topologies (Public/Private Subnets, VPCs, Internet Gateways, NAT Gateways).
- **Real-Time Queuing & Latency Simulation**: Deterministic and stochastic performance modeling using the M/M/c (Erlang-C) queueing model, computing queue wait times (W_q), queue depths (L_q), and latency degradation under load.
- **Serverless & Container Cold Start Emulation**: Models cold start penalties for AWS Lambda, Aurora Serverless, and container pull latencies for ECS Fargate and EKS pods with provisioned concurrency controls.
- **Multi-AZ Failure & Blast Radius Analysis**: Simulate instant Availability Zone outages to observe live traffic redistribution, remaining capacity stress, and cascading queue collapse.
- **Live Cloud Cost & FinOps Estimation**: Real-time monthly bill calculation with tier-based pricing (compute instances, provisioned IOPS, egress bandwidth, managed database licenses).
- **Infrastructure-as-Code (IaC) Synthesis**: Native 1-click compilation of visual diagrams into clean, validated Terraform HCL (AWS, Azure, GCP), AWS CloudFormation (YAML/JSON), Kubernetes Helm Charts, Pulumi (TypeScript/Python), and Docker Compose.
- **Live Cloud Discovery & Drift Detection**: Bi-directional reconciliation engine capable of importing live AWS infrastructure via AssumeRole credentials and scanning existing Terraform state files.

## Mathematical Simulation Models

ArchViz models system performance using rigorous queueing theory:

1. **Erlang-B Blocking Recurrence (Numerically Stable)**:
   - Formula: `B(0, A) = 1`, `B(k, A) = (A * B(k-1, A)) / (k + A * B(k-1, A))`
   - Parameters: `A = λ / μ` (Traffic Intensity in Erlangs), `λ` = Request arrival rate, `μ` = Service rate per node.

2. **Erlang-C Delay Probability (M/M/c Queue)**:
   - Formula: `C(c, A) = (c * B(c, A)) / (c - A * (1 - B(c, A)))`
   - Parameters: `c` = Number of parallel servers/instances, `ρ = A / c` = System utilization factor (for `ρ < 1`).

3. **Average Queue Wait Time (W_q)**:
   - Within Capacity (`ρ < 1`): `W_q = C(c, A) / (c * μ - λ)`
   - Overloaded Capacity (`ρ >= 1`): `W_q = min(10 * (1 / μ), (1 / μ) * ρ^2)`

4. **Average Queue Depth (L_q, Little's Law)**:
   - Formula: `L_q = λ * W_q`

5. **Network Latency & Geo-Routing**:
   - Intra-AZ network hop: ~1 ms
   - Cross-AZ network hop: ~2–5 ms
   - Cross-Region inter-continental routing: ~35–85 ms
   - Compliance Geo-fencing hop overhead: Strict EU (+45 ms), Strict US (+35 ms)
   - Third-Party API Latency Bottlenecks: OpenAI API (+800 ms), Stripe (+120 ms), Twilio (+60 ms)
   - Cold Starts: Serverless Lambda (1200 ms default scaled with memory), Container pull (3000 ms base)
   - CPU Burst Throttling: When T3/T4g burst credits reach 0, throttled instances incur `(5 * baseLatency) + 2000 ms` penalty.

## Supported Infrastructure-as-Code (IaC) Formats

- **Terraform HCL**: Multi-cloud HCL modules with variables, outputs, and security groups across AWS (`aws_*`), Azure (`azurerm_*`), and Google Cloud (`google_*`).
- **AWS CloudFormation**: CloudFormation templates in JSON and YAML formats with parameters, conditions, and resource cross-referencing.
- **Kubernetes Helm Charts**: Production Helm charts containing `Chart.yaml`, `values.yaml`, and manifest templates (Deployments, Services, ConfigMaps, Ingress).
- **Pulumi**: Programmatic IaC in TypeScript and Python using official Pulumi AWS/Azure/GCP SDKs.
- **Docker Compose**: Local multi-container development configurations with service ports, environment configs, and bridge networks.

## Documentation & Canonical Endpoints

- [ArchViz Studio Application](https://archviz-studio.vercel.app/): Main interactive canvas and architecture design environment.
- [Public Architecture Gallery & Templates](https://archviz-studio.vercel.app/templates): Pre-built enterprise architectures (Serverless Web App, High-Availability Multi-AZ Microservices, Data Pipeline, E-commerce).
- [Cloudcraft vs ArchViz Comparison](https://archviz-studio.vercel.app/compare/cloudcraft): In-depth comparison highlighting real-time simulation, multi-cloud IaC export, and pricing.
- [Terraform Architecture Generator Tool](https://archviz-studio.vercel.app/tools/terraform-visualizer): Visual canvas to Terraform HCL converter with live syntax highlighting and download.
- [Deep Technical Specification & Component Catalog](https://archviz-studio.vercel.app/llms-full.txt): Comprehensive technical specification for LLM code generation and RAG ingestion.
- [XML Sitemap](https://archviz-studio.vercel.app/sitemap.xml): Full URL crawl index.