Open-source alternatives to Tokio Uring

Arrow is a functional programming library for Kotlin that provides tools for implementing data-oriented programming patterns. It serves as a framework for typed error handling, a concurrency toolkit, and a library for the manipulation of immutable data. The project distinguishes itself through specialized capabilities for managing application failures using explicit types instead of exceptions and implementing resilience patterns such as circuit breakers and retry policies for distributed services. It also provides optics to update and query deeply nested immutable data structures without man

This repository is a collection of practical code snippets and implementation patterns for Flutter and Dart. It serves as a comprehensive guide and reference for asynchronous programming, state management patterns, and UI component design. The project provides advanced language reference material covering generics, reflection, factory constructors, and null-aware operators. It also includes specific utilities for manipulating Dart collections, such as helper methods for transforming and filtering maps, lists, and iterables. The coverage extends to high-level capabilities including asynchrono

This project is a PHP implementation of the Promises/A+ specification, providing a library for managing asynchronous operations and deferred values. It serves as an asynchronous task coordinator that allows for the creation of non-blocking code through a promise-based pattern. The library enables the simulation of asynchronous coroutines, allowing non-blocking code to be written in a linear style. It features duck-typed interoperability, which allows it to integrate with any foreign object that implements a then method regardless of class inheritance. The project covers broader capabilities

liburing is a C library that provides a low-level wrapper for the Linux kernel io_uring interface. It serves as a programming interface for executing non-blocking disk and network operations and abstracts the system calls required to communicate with the Linux kernel. The library focuses on reducing system call overhead and context switching for high-throughput data processing. It implements mechanisms for shared ring buffers, zero-copy buffer registration, and fixed-file descriptor mapping to minimize internal lookup and reference counting overhead. The project covers asynchronous input and

Safetensors is a secure tensor serialization format and library designed for storing and distributing model weights. Its primary purpose is to provide a safe file format for machine learning tensors that prevents the execution of arbitrary or malicious code during the deserialization process. The project is distinguished by its use of zero-copy memory mapping, which reads data from disk directly into memory to minimize overhead. It enables cross-framework compatibility, allowing tensor data to be serialized and deserialized across different machine learning libraries. The system covers high-

This project is a Go library and runtime for loading and managing eBPF programs and maps. It provides a bytecode loader and kernel interface to inject instructions into kernel hooks for system-level execution and observability across both Linux and Windows operating systems. The library features a relocation engine and tooling to ensure program compatibility across different kernel versions and distributions. It supports portable deployment by embedding compiled objects for multiple CPU architectures into a single binary and provides the ability to load signed system drivers on Windows. The

freecache is a high-performance, concurrent in-memory key-value store library for Go applications. It functions as a fixed-size cache that manages large volumes of data while eliminating garbage collection overhead through the use of preallocated memory and off-heap storage. The system maintains a predictable memory footprint by utilizing fixed memory allocation and circular buffer management, where new entries automatically overwrite the oldest data upon reaching capacity. It further reduces contention between simultaneous readers and writers through lock-striped concurrent access. The libr

CapnProto is a zero-copy serialization framework and remote procedure call system. It serves as a C++ communication library providing a schema-based data interchange format that eliminates the need to encode or decode data before reading it from memory. The system enables high-performance data serialization and low-latency network communication. It supports cross-language data exchange by using a defined schema to ensure consistent binary representation across different platforms. The framework provides tools for implementing remote procedure calls, allowing functions to be invoked on a remo

async-std is a Rust asynchronous runtime and non-blocking I/O library. It serves as an asynchronous standard library, providing a set of alternatives to the core Rust library for managing concurrent tasks, networking, and file system access. The project implements a standard-library-mirrored API, offering asynchronous versions of existing synchronous types to maintain a familiar interface. This approach allows for asynchronous Rust development using an interface that reflects the ergonomics of the language's standard library. The runtime includes a task executor for scheduling and executing

Async-task is a low-level Rust concurrency library and asynchronous task executor designed for building custom asynchronous runtimes and managing background task execution pipelines. It provides concurrency primitives that handle atomic task state management, heap-allocated futures, output retrieval, and type-erased execution handles. The library coordinates task execution, cancellation, and polling states across multiple threads using lock-free atomic operations and reference counting. Developers can supply custom scheduling logic and metadata hooks that dictate when and how tasks are queue

Smol is a lightweight Rust asynchronous runtime and task executor. It provides a suite of foundational tools for scheduling and executing asynchronous futures, managing background workloads, and handling non-blocking network and file operations. The project includes a compatibility layer for adapting asynchronous types and execution contexts, enabling futures to run across different runtime ecosystems. It also features a mechanism to offload synchronous I/O and CPU-intensive operations to dedicated worker thread pools to prevent runtime stalls. Its capabilities cover non-blocking network I/O

This library is an Arduino-compatible client for the messaging protocol designed for resource-constrained embedded systems. It provides a messaging interface that enables microcontrollers to establish persistent connections with brokers to exchange data and control signals. The library distinguishes itself through a focus on predictable resource usage, utilizing manual memory buffer management and a state-machine approach to handle connection persistence and automatic reconnections. It operates on a stream-oriented network abstraction, allowing it to interface with various hardware transport

Sled is an embedded key-value store and ACID-compliant database designed for high-performance data persistence. It functions as a log-structured storage engine that organizes data using B+ trees to support efficient range queries and prefix scans. The engine implements a zero-copy data store model, utilizing epoch-based reclamation to provide direct references to cached values without memory allocations. It distinguishes itself through a combination of write-ahead logging, page cache optimizations to reduce write amplification on flash storage, and serializable transactions for atomic multi-k

Greenlet is a multi-threaded asynchronous executor and web worker task runner. It serves as an execution wrapper that offloads heavy computation to background workers to maintain user interface responsiveness and prevent the main execution thread from blocking. The project includes a transferable object handler designed to move large data buffers between threads. This mechanism avoids the overhead of memory copying to optimize data transfer. The library covers the management of web workers, facilitating asynchronous task isolation and main thread offloading.

Zero-cost asynchronous programming in Rust

ProcMon-for-Linux is an eBPF-based system observability tool and process monitor for Linux. It functions as a system call tracer and activity logger, capturing real-time kernel and user-space events to analyze operating system behavior. The project features a text user interface for inspecting recorded trace files. It separates high-performance headless event recording from the analysis interface to prevent data loss during heavy system loads. The tool provides capabilities for system call tracing and activity monitoring, including the ability to filter events by process identifiers or speci

mQuickjs is a compact JavaScript engine designed for memory-constrained devices, such as microcontrollers, requiring as little as 10 KB of RAM and 100 KB of ROM. It achieves this through a fixed-buffer memory allocation scheme that avoids system heap usage, a compacting tracing garbage collector that eliminates fragmentation, and a ROM-embedded standard library that reduces both memory footprint and startup time. The engine compiles JavaScript source into portable bytecode that can be stored in persistent memory and executed without reparsing, and it emulates IEEE 754 floating-point arithmetic

Meilisearch is a Rust-based search engine providing typo-tolerant full-text and vector-based semantic search with real-time conversational capabilities.

simdjson is a high-performance, header-only C++ library designed for parsing, querying, and serializing JSON data with minimal memory overhead. It functions as a hardware-aware data processing engine that leverages vector instructions to achieve gigabyte-per-second parsing speeds. By detecting host processor capabilities at runtime, the library automatically selects the most efficient instruction sets to accelerate structural analysis and validation. The library distinguishes itself through a focus on extreme efficiency and resource management. It utilizes memory mapping and padded buffer ali

This project is a compatibility layer and abstraction library for building native C++ addons for Node.js. It provides a consistent C++ API that allows a single codebase to compile against multiple versions of the V8 JavaScript engine by hiding version-specific differences behind macros and inline functions. The toolkit differentiates itself through specialized wrappers and managers, including a monadic wrapper for handles to avoid explicit null checks, a persistent reference manager for objects that must survive handle scope disposal, and a buffer abstraction API for version-compatible memory

Kore is an event-driven web and WebSocket server framework designed for building high-performance web services and APIs in C or Python. It functions as a secure, concurrent application framework that utilizes non-blocking I/O and isolated worker processes to handle high-concurrency traffic. The project is distinguished by a security-first architecture that features OS-level process sandboxing, privilege separation, and the isolation of private encryption keys into dedicated processes. It enables zero-downtime deployments through dynamic module hot-reloading and provides automated TLS certific

OpenVINO is an AI inference engine and model serving platform designed to execute optimized deep learning models across CPUs, GPUs, and NPUs through a unified API. It includes a model optimization toolkit for converting, quantizing, and compressing models from various frameworks, alongside a specialized generative AI runtime for large language models. The project distinguishes itself through a plugin-based hardware acceleration layer that maps neural network operations to vendor-specific drivers. It features advanced execution mechanisms such as continuous batching, speculative decoding, and

Kitex is a high-performance remote procedure call framework for building distributed microservices in Go. It provides a communication system compatible with gRPC and Protobuf standards and includes a suite of tools for microservices governance. The framework features proxyless service mesh integration, which allows it to manage network traffic and governance directly without adding sidecar proxy latency. It optimizes data throughput through a high-performance communication library that minimizes memory copies. The system covers a broad surface of distributed capabilities, including service d

Hypernova is a distributed rendering engine and server-side rendering service that transforms JavaScript components into HTML on a remote server. It functions as a web page hydration framework, allowing interactive behavior to be attached to static server-generated markup by deserializing application state in the browser. The system utilizes an isolated JavaScript runtime that executes components within sandboxed virtual machines to prevent global state leakage and interference between concurrent requests. It employs a distributed architecture to execute rendering tasks across multiple worker

pysheeet is a technical reference library providing a curated collection of code snippets and implementation patterns for advanced Python development, system integration, and high-performance computing. It serves as a comprehensive guide for implementing low-level network programming, native C extensions, and asynchronous and concurrent programming. The project provides specialized frameworks for the development and deployment of large language models, including tools for distributed GPU inference and high-performance serving. It also includes detailed patterns for high-performance computing

Fury is a multi-language binary serialization framework designed for encoding domain objects and complex graphs to facilitate cross-language data exchange. It includes an interface definition language compiler that translates schema definitions into idiomatic native types and serialization boilerplate across multiple languages. The project distinguishes itself through a zero-copy binary reader that allows specific fields to be accessed without deserializing the entire object, as well as an object graph serializer that preserves circular references and referential integrity. It also features a

cuml is a GPU-accelerated machine learning library and framework that uses CUDA to accelerate tabular data preprocessing and model execution. It provides a suite of tools for training and deploying classification, regression, and clustering models on NVIDIA GPUs and GPU clusters. The library is designed for scalability, offering a distributed GPU machine learning environment that can spread computation and data across multiple hardware accelerators and nodes to handle datasets exceeding single-device memory. It mirrors standard estimator interfaces to allow the replacement of CPU-based models