awslabs/llrt
Llrt
llrt is a low-latency JavaScript runtime based on the QuickJS engine, specifically designed for executing asynchronous functions in serverless environments. It provides a lightweight execution layer optimized for fast startup times and minimal memory usage when running ES2023 workloads.
The project differentiates itself by bundling natively optimized cloud service SDKs directly into the runtime binary to eliminate external dependency loading. To further reduce cold start latency, it implements parallel connection warming for TLS and network handshakes during the startup phase.
The runtime covers a broad range of capabilities including network connectivity via HTTP and UDP, filesystem management, and modular cryptographic operations. It also includes memory management tools such as configurable garbage collection thresholds and a set of web API polyfills to ensure compatibility with standard JavaScript libraries.
The project includes a built-in test runner for local function verification and supports deployment via infrastructure as code templates.
Features
- Low-Latency JavaScript Execution - Executes JavaScript applications in a lightweight environment specifically optimized for low-latency serverless execution.
- JavaScript Runtimes - Provides a lightweight JavaScript runtime based on QuickJS optimized for efficiency and fast startup.
- Native SDK Integration - Bakes high-performance cloud service clients directly into the runtime binary to eliminate external dependency loading.
- Serverless Execution Environments - Provides an optimized runtime environment for event-driven, ephemeral code execution across various architectures.
- Cloud Native Application Development - Provides a runtime optimized for building and executing cloud-native serverless applications.
- Native SDK Binaries - Bakes natively optimized cloud service SDKs directly into the runtime binary to remove dependency loading overhead.
- Serverless Function Executions - Runs asynchronous handler functions to process events in a low-latency serverless environment.
- Function Warm-up Utilities - Initializes network connections in parallel during startup to eliminate cold-start latency for initial requests.
- Native Binary Packagers - Bundles cloud service clients into the runtime binary as part of the native binary packaging to eliminate cold starts.
- Parallel Connection Warming - Initializes TLS and network connections in parallel during startup to remove handshake latency from the first request.
- Cloud Native Runtimes - Acts as an execution layer that bundles native cloud clients and network optimizations to accelerate function initialization.
- Cold Start Optimizations - Reduces cold start latency through parallel connection warming and the use of lightweight binaries.
- Connection Warming - Reduces TLS latency by initializing secure connections in parallel during the runtime startup phase.
- Asynchronous Event Loops - Implements a custom event-driven execution model designed specifically for low-latency serverless workloads.
- QuickJS-Based Runtimes - Uses the QuickJS engine to provide a lightweight environment with fast startup and low memory overhead.
- Low-Latency Execution - Optimizes JavaScript execution for minimal memory usage and fast startup times in serverless environments.
- Serverless Workload Execution - Executes ES2023 JavaScript workloads with a focus on minimal memory footprint and low latency.
- Configurable Garbage Collection - Implements configurable garbage collection thresholds to prevent memory exhaustion in constrained serverless environments.
- Cloud Service SDKs - Bundles specific cloud service clients and endpoints into the runtime to reduce initialization time.
- Local Function Execution - Provides utilities to run serverless functions locally using mock events to emulate cloud environments.
- Native SDK Bundling - Bakes native implementations of cloud service clients into the runtime to reduce cold starts and binary size.
- AWS API Clients - Integrates high-performance AWS API clients directly into the runtime to reduce cold start latency.
- Application Cloud Deployments - Supports packaging application code into container images for deployment to cloud hosting environments.
- Infrastructure Deployment - Includes dedicated construct libraries to automate the deployment of serverless functions via infrastructure as code.
- Serverless Deployment - Provides tools to package and deploy code to event-driven serverless cloud environments.
- Garbage Collection - Implements configurable garbage collection thresholds to reclaim memory and prevent application crashes in serverless environments.
- Memory-Threshold GC Tuning - Allows configuring memory limits that force the garbage collector to run and prevent memory exhaustion.
- Embedded JavaScript Interpreters - Utilizes an embedded QuickJS interpreter to provide a high-performance, small-footprint environment.
- JavaScript and Web-Standard - Supports development of small-footprint applications using ES2023 and essential web-standard APIs.
- API Polyfills - Provides embedded polyfills for standard web APIs to ensure compatibility with common JavaScript libraries.
- Resource Limit Configurations - Allows configuration of memory limits and idle socket timeouts to ensure stable system performance.
- Function Monitoring - Provides a CLI to fetch and filter execution logs from deployed serverless functions.
- Memory Usage Analyzers - Computes detailed memory statistics, including object and string counts, to monitor runtime resource usage.
- Garbage Collection Triggers - Forces garbage collection when a configurable memory threshold is reached to prevent crashes in constrained environments.
- Built-in Testing Frameworks - Includes an integrated, lightweight testing framework for verifying functions with standard assertions.
