GitHub: dbos_experiments/exp6

Experiment 6

Summary

This experiment explores DBOS scope management, class-based workflows, caching patterns, application lifecycle, and Docker containerization. It demonstrates different aspects of DBOS application architecture through multiple specialized implementations.

Files Description

Core Application Files

• scope.py - DBOS class-based workflow demonstration:
  • @DBOS.dbos_class() decorator usage for class-based workflows
  • Instance-level state management within workflows
  • Step execution within class instances
  • Demonstrates how DBOS handles object scope in workflows
• fastapi_scope.py - FastAPI integration with DBOS classes:
  • Class-based workflow with instance state (counter)
  • Synchronous and asynchronous workflow triggers
  • FastAPI endpoints for workflow management
  • Demonstrates object lifecycle in web applications
• fastapi_scope_cache.py - Caching patterns and application lifecycle:
  • Global cache implementation (credentials_cache)
  • Application lifespan management with @asynccontextmanager
  • DBOS lifecycle hooks (DBOS.launch(), DBOS.destroy())
  • Credential caching and reuse patterns
  • User processing workflows with cached data
• fastapi_scope_shutdown.py - Application shutdown and scheduled workflows:
  • Fibonacci calculation in steps (CPU-intensive operation)
  • Queue-based processing with 50 concurrent steps
  • Scheduled workflows (every 5 seconds)
  • Application shutdown endpoint
  • Process lifecycle management

• integration.py - Integration patterns (details would need file inspection)

• nested_wfs.py - Nested workflow patterns (details would need file inspection)

• scope_funcs.py - Scope utility functions (details would need file inspection)

Configuration Files

• __init__.py - Python package initialization file

• dbos-config.yaml - DBOS configuration for database and application settings

• envrc-template - Environment variables template for direnv integration

• requirements.txt - Python dependencies for the experiment

Docker Configuration

• Dockerfile - Container configuration for DBOS application:
  • Based on python:3.12-slim
  • Installs minimal dependencies: dbos, fastapi, psycopg2-binary, python-json-logger
  • Configures environment variables for PostgreSQL connection
  • Exposes port 8000 for FastAPI
  • Runs fastapi_scope_shutdown.py by default
  • Demonstrates containerized DBOS deployment

Cache Files

• __pycache__/ - Python bytecode cache directory

Key Concepts Demonstrated

DBOS Class System

• Object Scope: How DBOS maintains object state across workflow execution
• Instance Management: Creating and managing class instances within workflows
• State Persistence: How instance variables persist during workflow execution

Application Lifecycle

• Startup/Shutdown: Proper DBOS initialization and cleanup
• Lifespan Management: FastAPI lifespan events integration
• Resource Management: Cache initialization and cleanup

Caching Patterns

• Global Caching: Application-level cache for credentials and configuration
• Scope Isolation: How cache state interacts with workflow scope
• Performance Optimization: Reducing redundant operations through caching

Containerization

• Docker Integration: Running DBOS applications in containers
• Environment Configuration: Database connection through environment variables
• Production Deployment: Container-ready DBOS application setup

DBOS

dbos start

Recent changes

• 2025-08-10

  bdd71b7

  added AI READMEs

• 2025-07-18

  70c5044

  added dbos 1.7

• 2025-02-21

  022d4ee

  scope experiments

• 2025-02-19

  a7f095e

  added example for AWS s3”