DotCompute Articles

Comprehensive guides, tutorials, and conceptual documentation for DotCompute.

Getting Started

• Installation Guide - Install DotCompute and configure your first project
• Quick Start Tutorial - Write your first GPU-accelerated kernel in 5 minutes
• Hello World Example - Complete walkthrough of a simple compute kernel

Architecture

Understanding DotCompute's design and component relationships:

• Architecture Overview - High-level system architecture
• Core Orchestration - Compute orchestration and service integration
• Backend Integration - How backends plug into the framework
• Memory Management - Unified memory system design
• Debugging System - Cross-backend validation architecture
• Optimization Engine - Adaptive backend selection design
• Source Generators - Compile-time code generation system

Developer Guides

Practical guides for common development scenarios:

• Kernel Development - Writing efficient compute kernels
• Backend Selection - Choosing and configuring backends
• Performance Tuning - Optimizing kernel performance
• Debugging Guide - Debugging kernels and troubleshooting issues
• Native AOT Compilation - Building with Native AOT for maximum performance
• Multi-GPU Programming - Scaling across multiple GPUs
• Memory Management Best Practices - Efficient memory usage patterns
• Dependency Injection Integration - Using DotCompute with DI containers
• Telemetry and Monitoring - Observability and performance monitoring
• Troubleshooting Common Issues - Solutions to common problems

Performance

Understanding and optimizing performance:

• Performance Characteristics - Expected performance for different operations
• Benchmarking Guide - How to benchmark your kernels
• Optimization Strategies - Advanced optimization techniques
• Hardware Requirements - Hardware considerations for different backends

Advanced Topics

Deep dives into specialized areas:

• SIMD Vectorization - CPU SIMD optimization techniques
• CUDA Programming - CUDA-specific features and optimization
• Metal Shading Language - Writing Metal compute shaders
• OpenCL Development - Cross-platform OpenCL programming
• Plugin Development - Creating custom backend plugins
• Analyzer Development - Writing custom Roslyn analyzers

Examples

Complete, runnable examples:

• Basic Examples - Simple kernels and common patterns
• Intermediate Examples - Real-world compute scenarios
• Advanced Examples - Complex multi-kernel pipelines

Migration Guides

Migrating from other compute frameworks:

• From ILGPU - Migrating ILGPU code to DotCompute
• From TorchSharp - Migrating ML tensor operations
• From Alea GPU - Migrating F# GPU code

Reference

Additional reference material:

• Diagnostic Rules (DC001-DC012) - Complete analyzer diagnostic reference
• Kernel Attribute Reference - [Kernel] attribute options
• Configuration Reference - All configuration options
• Glossary - Terminology and definitions

Contributing

Help improve DotCompute:

• Contributing Guide - How to contribute to the project
• Code of Conduct - Community guidelines
• Development Setup - Setting up a development environment