This article examines Log4j2 asynchronous logging bottlenecks, explains the underlying Disruptor queue mechanics, identifies root causes of thread blocking, and presents practical strategies—including queue tuning, log classification, bytecode instrumentation, and IDE plugins—to achieve fine‑grained, performant log control.
pfinder, JD’s in‑house APM system, provides full‑link monitoring, multi‑dimensional metrics, automatic instrumentation, topology mapping, trace analysis, AI‑driven fault detection by leveraging bytecode enhancement techniques such as ASM, Javassist, ByteBuddy and ByteKit, and integrates with JVM agents for hot‑deployment and trace propagation.
This article provides a comprehensive technical overview of JD's self‑built PFinder APM system, detailing its core concepts, multi‑dimensional monitoring capabilities, bytecode‑enhancement mechanisms using ASM, Javassist, ByteBuddy and ByteKit, JVMTI‑based agents, service and plugin loading, trace‑ID propagation across threads, and a prototype hot‑deployment solution.
This article presents a comprehensive approach to reducing thread overhead in Android applications by detecting, counting, and consolidating thread creation through bytecode instrumentation, proxy thread pools, and stack size trimming, ultimately improving memory usage, CPU load, and UI smoothness.
This article explains the background, principles, loading mechanisms, usage workflow, testing considerations, and advantages of JavaAgent technology for dynamic Java bytecode instrumentation and runtime protection.
This article explains the principles of Interactive Application Security Testing (IAST), details the JavaAgent-based agent instrumentation approach, and provides a step‑by‑step guide to building a Java IAST agent using Javassist, custom class loaders, ThreadLocal tracing, and Maven packaging.
This article details how Huolala built a comprehensive full‑chain monitoring and tracing platform, covering the historical evolution of observability tools, the company’s multi‑stage monitoring architecture, bytecode‑enhanced instrumentation, trace sampling strategies, and a "what‑you‑see‑is‑what‑you‑get" visualization approach.
This article explains how to use Java bytecode instrumentation with ASM and a JavaAgent to automatically monitor, capture, and report exceptions across a backend system, covering exception fundamentals, best‑practice handling, and practical implementation steps.
This article explores how Alibaba's Eagleeye and the open‑source SkyWalking implement distributed tracing, covering background challenges, Dapper concepts, design goals, data models (Trace, Segment, Span), unique ID schemes, instrumentation techniques, data collection, storage mechanisms, and transmission strategies.
This article explains how to build a full‑tracking solution for Android by creating a Gradle plugin that registers a Transform, leveraging the Transform API and ASM bytecode instrumentation to automatically inject analytics code into click events and lifecycle callbacks, while covering the underlying concepts, implementation steps, and deployment process.
This article explains Java exception fundamentals, best‑practice handling, and demonstrates how to use ASM bytecode enhancement together with a JavaAgent to automatically capture, monitor, and report exception information in a non‑intrusive way for backend services.
vivo’s “Moonlight Treasure Box” platform records and replays production traffic using a customized Jvm‑Sandbox‑Repeater Java‑Agent, handling GC limits, thread‑pool tracing, data desensitization, deduplication, timestamp alignment, noise reduction and sub‑call matching, while providing plugins, visual call‑chain analysis and coverage metrics that have integrated 200 services, executed over 10 000 sessions, and boosted testing efficiency by more than 80 %.
This article describes the design and implementation of a Gradle plugin that extends JaCoCo to collect incremental code‑coverage data for Android applications, covering project background, significance, JaCoCo injection mechanisms, compile‑time instrumentation, runtime data handling, report generation, and practical results.
The 2017‑initiated vivo distributed tracing system leverages a JavaAgent‑based micro‑kernel architecture, using ByteBuddy for non‑intrusive bytecode instrumentation, a Disruptor lock‑free queue, and Kafka to capture Trace/Span data—including cross‑thread propagation—while employing sampling, degradation, and JVM metrics to ensure 94% adoption stability.
This article describes Ctrip's practical deployment of OpenRASP‑based IAST, outlines the challenges of data pollution caused by traffic replay, and presents a Java bytecode instrumentation solution that intercepts SocketOutputStream writes to prevent dirty data from persisting in databases, caches, and message queues.
This article explains how to build a Java Agent that leverages bytecode instrumentation, class‑loader mechanisms, and AOP techniques to provide lightweight, configurable performance profiling and monitoring for Java services, covering requirements, technical foundations, implementation steps, and testing procedures.
JVM‑Sandbox, an open‑source real‑time, non‑intrusive bytecode‑enhancement framework developed by Alibaba’s Technical Quality team since 2016, provides dynamic AOP, modular management, and HTTP‑based control to support fault injection, dependency analysis, recording/replay, and precise regression, dramatically improving testing efficiency and stability for large‑scale services.
This article explores the causes of UI stutter in Android apps, explains how frame rate and dropped frames affect perceived smoothness, and details practical profiling solutions—including TraceView, BlockCanary, ArgusAPM, and Matrix‑TraceCanary—along with implementation techniques for accurate lag detection and mitigation.
This article details the design, technology selection, and implementation of WMDA, a no‑tracking (no‑annotation) user‑behavior analysis SDK for Android, covering bytecode‑instrumentation via a Gradle plugin, event collection for activities, fragments, and clicks, as well as processing, storage, reporting, and current limitations.
Facebook tackled Android performance monitoring by replacing manual instrumentation with a rule‑based bytecode rewriter that automatically injects tracing markers, minimizes overhead, and captures cross‑thread activity, offering a scalable solution for real‑world telemetry collection.
