Workerd

workerd is a serverless edge runtime designed for executing lightweight, distributed functions at the network edge. It utilizes a V8-based JavaScript engine to provide fast startup and low memory overhead, while maintaining a WebAssembly-compatible execution environment that allows modules to run alongside JavaScript for high-performance computational tasks.

The runtime supports isolate-based multi-tenancy to run multiple independent execution contexts within a single process. It implements an event-driven execution model that triggers code based on network requests or scheduled events and includes support for privileged socket inheritance to operate under unprivileged user accounts.

The project covers a broad set of capabilities including serverless API development, AI inference deployment using GPU hardware and vector databases, and automated browser orchestration for web scraping. Additional functionality encompasses global state management via SQL databases and key-value stores, background job scheduling with message queues, and the delivery of static assets through a content delivery network.

Development is supported by a command-line interface for project management, custom build pipelines, and tools for pinning runtime behavior to specific dates to ensure consistency.

Features

Edge Computing Platforms - Implements a distributed platform that executes code closer to end users to minimize request latency.

JavaScript Runtimes - Implements a high-performance V8-based runtime for executing server-side JavaScript and WebAssembly modules.

Inference and Deployment - Deploys and executes large language models and image generation tasks using serverless GPU hardware.

Edge Data Persistence - Maintains SQL databases, key-value stores, and object storage for global data access and coordination.

Edge Function Runtimes - Provides a globally distributed runtime for executing lightweight serverless functions at the network edge.

Serverless Execution Models - Employs an event-driven model that triggers code execution based on network requests or scheduled events.

Edge AI Inference - Provides serverless GPU hardware to execute large language models and image generation tasks at the network edge.

Serverless Backend Hosting - Hosts scalable backend APIs and logic with optimized global placement to minimize request latency.

Embedded Wasm Runtimes - Provides an embedded Wasm runtime that allows WebAssembly modules to run alongside JavaScript.

V8 Engine Integrations - Integrates the V8 engine to provide high-performance execution of JavaScript and WebAssembly in a server-side environment.

Serverless Engine Optimizations - Utilizes a V8-based engine optimized for fast startup and low memory overhead in serverless contexts.

Isolate-Based Multi-Tenancy - Runs multiple independent execution contexts in a single process to reduce memory overhead and startup latency.

Wasm Execution Frameworks - Provides a framework that utilizes WebAssembly to run high-performance computational logic in serverless environments.

Static Asset Hosting - Delivers static files via a global content delivery network and cache to reduce rendering latency.

Global State Coordination - Coordinates data access across distributed nodes using integrated key-value stores and vector databases.

Semantic Search Engines - Integrates vector databases to enable AI-powered search based on the conceptual meaning of data.

Background Job Schedulers - Schedules recurring tasks and long-running operations using automatic retries and message queues.

Browser Automation - Runs programmatic serverless browser instances to interact with and scrape web pages.

Headless Browser Orchestrators - Provides a system to manage and control isolated headless browser instances for web scraping at scale.

cloudflareworkerd

Features

Star history