Skip to content

README.md

Latest commit

 

History

History
114 lines (87 loc) · 3.88 KB

README.md

File metadata and controls

114 lines (87 loc) · 3.88 KB

01. Ray Core - Distributed Computing Fundamentals

Welcome to Ray Core! This section covers the core concepts and building blocks of Ray for distributed computing. If you're new to Ray, start here to build a solid foundation.

What is Ray Core?

Ray Core is the foundation of the Ray framework, providing simple primitives for building and running distributed applications:

  • Tasks - Stateless functions executed remotely
  • Actors - Stateful classes running in the cluster
  • Objects - Immutable values stored in Ray's distributed object store

Directory Structure

01-ray-core-distributed-computing/
├── notebooks/                    # Main learning notebooks
│   ├── Ray_Core_1_Remote_Functions.ipynb    # Remote tasks and parallel execution
│   ├── Ray_Core_2_Remote_Objects.ipynb      # Distributed object store
│   ├── Ray_Core_3_Remote_Classes_part_1.ipynb   # Actors introduction
│   ├── Ray_Core_4_Remote_Classes_part_2.ipynb   # Advanced actor patterns
│   ├── Ray_Core_5_Best_Practices.ipynb      # Best practices and patterns
│   └── Ray_Core_6_Compiled_Graphs_experimental.ipynb  # Experimental features
├── bonus/                        # Additional examples
│   ├── 1_core_examples.ipynb     # Extra Ray Core examples
│   └── 2_task_lifecycle_deep_dive.ipynb  # Deep dive into task lifecycle
├── solutions/                    # Exercise solutions
│   ├── ex_01_solution.ipynb
│   ├── ex_02_solution.ipynb
│   ├── ex_03_solution.ipynb
│   └── ex_04_solution.ipynb
├── utils/                        # Helper utilities
│   ├── model_helper_utils.py
│   └── tasks_helper_utils.py
└── README.md

Learning Path

1. Remote Functions (Tasks)

Notebook: Ray_Core_1_Remote_Functions.ipynb

Learn the fundamentals of Ray tasks:

  • Converting Python functions to distributed tasks with @ray.remote
  • Parallel execution patterns (Monte Carlo, Fibonacci)
  • Task dependencies and DAG patterns
  • Resource management and scheduling

2. Remote Objects

Notebook: Ray_Core_2_Remote_Objects.ipynb

Understand Ray's distributed object store:

  • ray.put() for storing objects
  • ray.get() for retrieving objects
  • Object references and futures
  • Efficient data sharing across tasks

3. Remote Classes (Actors) - Part 1

Notebook: Ray_Core_3_Remote_Classes_part_1.ipynb

Introduction to stateful computation with actors:

  • Creating actors with @ray.remote
  • Actor methods and state management
  • Actor handles and references

4. Remote Classes (Actors) - Part 2

Notebook: Ray_Core_4_Remote_Classes_part_2.ipynb

Advanced actor patterns:

  • Actor pools and load balancing
  • Async actors
  • Actor lifecycle management
  • Fault tolerance with actors

5. Best Practices

Notebook: Ray_Core_5_Best_Practices.ipynb

Production-ready patterns:

  • Anti-patterns to avoid
  • Memory management
  • Performance optimization
  • Debugging techniques

6. Compiled Graphs (Experimental)

Notebook: Ray_Core_6_Compiled_Graphs_experimental.ipynb

Explore experimental features for optimized execution graphs.

Prerequisites

  • Python 3.12+ installed
  • Ray (latest version) installed via uv (see main README)
  • Basic Python knowledge (functions, classes, decorators)

Getting Started

# Navigate to this directory
cd 01-ray-core-distributed-computing

# Start Jupyter
jupyter notebook

# Or use JupyterLab
jupyter lab

Start with notebooks/Ray_Core_1_Remote_Functions.ipynb and progress through the numbered notebooks.

Resources