0 views collected around this technical thread.
This article explains the concept of zero‑copy, compares traditional I/O copying with mmap, sendfile, DMA scatter/gather and splice techniques, and shows how Java NIO leverages mmap and sendfile to achieve high‑performance data transfer while minimizing CPU involvement.
This article provides a detailed overview of Netty, a high‑performance asynchronous NIO framework, covering its key features, architectural design, core components, high‑performance strategies, and typical use cases in internet services, gaming, and big‑data systems, while also highlighting why it outperforms native JDK NIO.
This article walks through Kafka’s network‑layer internals, detailing how the client prepares, pre‑sends, and actually sends messages, then reads and parses responses, with code snippets that illustrate the Selector, KafkaChannel, and related classes in Java NIO.
This article explains the historic C10K problem of handling ten thousand concurrent connections, compares traditional BIO with modern I/O models such as NIO, select, poll, epoll, and AIO, and provides Java and C++ code examples illustrating how each model improves scalability and performance.
This article deeply analyzes Kafka broker’s network architecture and request‑handling pipeline, walking through simple sequential models, multithreaded async designs, the Reactor pattern with Java NIO, key thread roles, core processing flow, and practical tuning parameters for high‑throughput, low‑latency deployments.
The article explains Java NIO’s non‑blocking channels, buffers, and selectors, then shows how the open‑source Tars RPC framework builds a multi‑reactor, multi‑thread network model on top of NIO—detailing server socket setup, event dispatch, session management, and read/write processing.
This article walks through a real‑world case of off‑heap memory leakage in the Dble proxy, detailing symptom observation, log analysis, monitoring, BTrace instrumentation, code review, and the eventual fix that delays buffer allocation until needed.
This article explains the zero‑copy technique in Java, covering I/O fundamentals, kernel read/write flow, mmap+write and Sendfile methods, virtual memory mapping, MappedByteBuffer and DirectByteBuffer usage, channel‑to‑channel transfer, Netty composite buffers, and how frameworks like RocketMQ and Kafka leverage zero‑copy for performance.
This article explains the concept of zero‑copy, how it eliminates redundant data copying in I/O operations, and demonstrates its implementation in Java through buffers, mmap+write, Sendfile, MappedByteBuffer, DirectByteBuffer, channel‑to‑channel transfers, and Netty’s composite buffers.
This article explains the concept of zero‑copy, its benefits for I/O performance, and how Java NIO, Netty, Kafka, RocketMQ and related APIs such as mmap+write and sendfile implement zero‑copy to avoid unnecessary data copying between kernel and user space.
This article summarizes the challenges and optimization techniques for building a high‑performance long‑connection service with Netty, covering non‑blocking I/O, Linux kernel tuning, client‑side testing, VM‑based scaling, data‑structure tweaks, CPU and GC bottlenecks, and the final results of achieving hundreds of thousands of connections and tens of thousands of QPS on a single server.
This article thoroughly examines Kafka's broker‑side network architecture, tracing its evolution from a simple sequential model to a high‑performance, event‑driven Reactor design using Java NIO, and provides practical tuning guidance for achieving optimal throughput and latency.
This article thoroughly analyzes Kafka broker's high‑throughput network architecture, tracing its evolution from simple sequential handling to a multi‑selector Reactor model, detailing Acceptor and Processor thread implementations, request‑handling pipelines, and practical tuning parameters for optimal performance.
This article explains how to build a scalable long‑connection server using Netty, discusses the underlying bottlenecks such as Linux kernel limits, CPU and GC issues, and provides practical code examples and tuning techniques to achieve hundreds of thousands of connections and high QPS.
This article explains how to build a scalable long‑connection server using Netty, covering the fundamentals of non‑blocking I/O, practical code examples for achieving millions of concurrent connections, common Linux kernel limits, CPU and GC bottlenecks, and a series of tuning techniques to dramatically improve QPS and stability.
This article explains the limitations of traditional blocking I/O, introduces Java NIO with selectors, compares the two models using a kindergarten analogy, critiques raw JDK NIO, and demonstrates how Netty simplifies and optimizes network development with concise server and client code examples while promoting a comprehensive Netty learning book.
This article explains how Netty abstracts Java NIO's event registration by using SelectionKey interestOps, detailing the registration of OP_ACCEPT and OP_READ events, the underlying doRegister and doBeginRead implementations, and the flow of events through the Netty pipeline with illustrative code snippets.
This article explains how to build a scalable long‑connection service using Netty, covering the underlying concepts, Linux kernel tuning, code examples for Java NIO and Netty, performance bottlenecks such as QPS, data‑structure and GC issues, and practical optimization techniques to achieve hundreds of thousands of concurrent connections.
This article explains how zero‑copy techniques such as mmap, sendfile, and splice reduce CPU involvement in data transfer by avoiding memory copies between kernel and user spaces, and shows how Java NIO and Netty implement these mechanisms for high‑performance backend I/O.
This article provides a comprehensive overview of Kafka performance optimization, covering network, disk, and complexity challenges, sequential write techniques, zero‑copy data transfer, page‑cache usage, reactor‑based networking, batch processing, compression, partition concurrency, and file‑structure design to help developers build high‑throughput, low‑latency streaming systems.