Welcome to the WVA documentation! This directory contains comprehensive guides for users, developers, and operators.
Getting started and using WVA:
- Installation Guide - Installing WVA on your cluster
- Configuration - Configuring WVA for your workloads
- CRD Reference - Complete API reference for VariantAutoscaling
- Multi-Controller Isolation - Running multiple WVA controller instances
- LeaderWorkerSet Support - Supporting LeaderWorkerSets as scale targets
Step-by-step guides:
- Quick Start Demo - Getting started with WVA
- Parameter Estimation - Estimating model parameters
- vLLM Samples - Working with vLLM servers
- GuideLLM Sample - Using GuideLLM for benchmarking
Integration with other systems:
- HPA Integration - Using WVA with Horizontal Pod Autoscaler
- KEDA Integration - Using WVA with KEDA
- Prometheus Integration - Custom metrics and monitoring
Understanding how WVA works:
- Modeling & Optimization - Queue theory models and optimization algorithms
- Controller Behavior - Event handling and reconciliation behavior
- Architecture Diagrams - System architecture and workflows
Contributing to WVA:
- Development Setup - Setting up your dev environment
- Testing - Running tests and CI workflows
- Agentic Workflows - AI-powered automation workflows
- Debugging - Debugging techniques and tools
- Contributing - How to contribute to the project
- Check the Troubleshooting Guide
- Open a GitHub Issue
- Join community meetings
Note: Documentation is continuously being improved. If you find errors or have suggestions, please open an issue or submit a PR!