A benchmark on an 8‑core Linux 6.6 system shows that switching Netty from epoll to io_uring lets a half‑sized thread pool achieve 3 % higher throughput, more than double per‑thread efficiency, and a 67 % reduction in CPU migrations, challenging the traditional rule of using twice‑the‑core thread counts.
This article analytically examines INFINI Pizza, a Rust‑implemented distributed search database, detailing its design philosophy, hierarchical data model, rolling‑partition‑shard architecture, share‑nothing + io_uring I/O stack, true real‑time indexing, in‑place partial updates, AI‑native hybrid search capabilities, ecosystem components, and a point‑by‑point comparison with Elasticsearch.
This article provides a comprehensive, step‑by‑step analysis of Linux's io_uring, covering its architecture, design principles, workflow, performance advantages over traditional models like epoll, practical C code examples, optimization techniques, real‑world use cases, and the challenges developers may face when adopting it.
The update introduces coroutine support for FTP and SSH clients, adds io_uring‑based HTTP server capabilities, expands RemoteObject and PDO drivers, fixes critical bugs in HTTP2, process handling, and memory management, and includes performance optimizations and important compatibility notes.
OpenSwoole 26.2.0 introduces full PHP 8.5 compatibility, native Fiber coroutine context via zend_fiber, an io_uring‑based reactor and async file I/O engine, Xdebug step‑debugging inside coroutines, detailed event‑loop latency metrics, and numerous bug fixes and breaking changes, with upgrade instructions for PECL, Docker, and Composer.
This article traces the evolution from select and poll to epoll and finally io_uring, explains how multishot accept and receive work, compares benchmark results, and provides practical guidance for migrating production servers from epoll‑based event loops to the newer io_uring model.
mmap, the classic Linux memory‑mapping technique, often surpasses the modern async io_uring in various I/O scenarios by eliminating redundant data copies, reducing system calls, and enabling zero‑copy access, while the article explains its fundamentals, workflow, performance comparisons, practical usage, pitfalls, and code examples.
This comprehensive guide explains why traditional I/O models become bottlenecks in high‑performance computing, introduces the modern io_uring framework with its submission and completion queues, walks through its design goals, core concepts, workflow, performance comparisons, optimization tips, real‑world use cases, and provides complete C examples for practical adoption.
Each new Linux kernel release introduces features like io_uring, DAMON, Landlock LSM, memfd_secret, and the IOMMU FD API that improve hardware interaction, memory efficiency, performance, and security, with detailed descriptions, real-world use cases, and examples for servers, embedded, multimedia, and real‑time systems.
This article explores the evolution of Linux I/O models—from blocking and non‑blocking to epoll—and introduces io_uring as a high‑performance asynchronous framework that reduces system‑call overhead, eliminates data copies, and unifies network and disk I/O for modern high‑concurrency applications.
This article traces the evolution of Linux I/O from blocking and non‑blocking models through epoll to the modern io_uring framework, explains its design goals, core concepts, system‑call workflow, and provides practical code examples and performance comparisons for high‑concurrency network and storage applications.
This article introduces Linux's io_uring framework, explains its design goals and advantages over traditional I/O models, details its core components and system calls, provides step‑by‑step implementation examples for file and network operations, and discusses performance comparisons and practical application scenarios.
This article provides a comprehensive overview of Linux's io_uring, explaining its design goals, shared‑memory mechanism, submission and completion queues, core system calls, performance advantages over traditional I/O models, typical use cases, and includes a complete example of a network server built with io_uring.
This article examines the integration of Linux io_uring into ByteDance's high‑performance Netpoll NIO library, detailing architectural changes, receive/send workflows, benchmarking methodology, and results that show over 10% higher throughput and 20‑40% lower latency while eliminating system calls.
The article surveys the evolution of POSIX abstractions from the 1970s to the present, explains why hardware trends and modern workloads expose limitations of the CPU‑centric model, and argues that a new set of higher‑level interfaces is needed for future systems.
Linux I/O models range from synchronous native operations to asynchronous kernel‑bypass techniques, with APIs such as open, read, write, mmap, sendfile and splice, while the modern io_uring interface—supporting interrupt, polling, and kernel‑polling modes via liburing—delivers superior performance, especially at high queue depths, compared to traditional libaio.
The new Btrfs async buffered‑write patch, developed by Meta engineer Stefan Roesch, boosts I/O throughput by more than two‑fold and cuts latency, with the changes slated for inclusion in the upcoming Linux 6.1 kernel release.
The article reviews high‑performance networking options—RDMA (including RoCE v2 and iWARP) and Linux’s io_uring—explaining their principles, hardware requirements, and benchmark results, and concludes that while RDMA delivers ultra‑low latency for specialized workloads, io_uring offers modest network benefits, leaving TCP as the default for most services.
Linux kernel 5.18, announced on May 22 by Linus Torvalds, brings a host of new features such as C11 compiler support, enhanced hardware compatibility, Btrfs ioctl extensions, io_uring upgrades, RSA crypto for virtio-crypto, and numerous driver and filesystem improvements.
This article explains how using the Storage Performance Development Kit (SPDK) optimizes virtualized I/O paths for cloud disks, covering its architecture, virtio vring basics, hot‑upgrade solutions, online migration techniques, and io_uring enhancements that together achieve up to 1.2 million IOPS with dramatically reduced latency.
