|
所在平台: Udemy |
课程主页: https://www.udemy.com/course/mastering-performance-analysis-with-perf/
课程评论：没有评论
**课程名称：** Mastering Performance Analysis with perf **课程概述：** 本课程是学习 Linux 官方性能分析工具 `perf` 的终极指南，旨在将您培养成为 Linux 系统性能分析的专家。课程历经两年精心打磨和多次修订，传授最前沿、最有效的 `perf` 使用技巧，助您深入洞察系统性能。 **课程内容亮点：** * **硬件事件监控 (Hardware Event Monitoring)：** 深入了解缓存未命中、分支预测错误等硬件事件，揭示硬件行为的细微之处。 * **系统事件监控 (System Event Monitoring)：** 实时监控上下文切换、缺页等关键系统事件，把握系统运行脉搏。 * **自顶向下微架构分析 (Top-Down Microarchitecture Analysis - TMA)：** 掌握此强大技术，撰写更高效的代码，最大限度地提升 CPU 性能。 * **火焰图 (Flame Graphs)：** 学会创建和解读火焰图，精准定位性能瓶颈。 * **Linux 跟踪框架 (Linux Tracing Frameworks)：** 探索 `tracepoints`、`kprobes`、`uprobes` 等跟踪工具，捕获深度的性能数据。 * **内核函数跟踪 (Kernel Function Tracking)：** 深入了解内核活动，快速识别段错误等故障点。 * **应用程序函数跟踪 (Application Function Tracing)：** 剖析应用程序行为，诊断并解决性能问题。 * **自定义性能分析 (Custom Performance Analysis)：** 学习设计自定义分析方法，例如计算函数延迟。 * **实操案例 (Hands-On Examples)：** 通过真实世界的案例练习，构建扎实的性能工程实践技能。 **学习目标：** 完成本课程后，您将能够自信地识别、诊断并优化应用程序和系统的性能问题，将这些技能融入到日常工作中。
Welcome to "Mastering Performance Analysis with perf"-the ultimate course to unlock the full power of Linux's official performance tool, perf. This course is designed to guide you to expert-level performance analysis, equipping you to become a true performance guru in Linux systems.Developed over two years with extensive revisions, this course delivers essential skills and techniques, ensuring you learn the most effective ways to leverage perf for deep system insights.Throughout the course, we'll cover a range of essential tools, methodologies, and technologies, including:Hardware Event Monitoring: Track events like cache misses and branch mispredictions for deeper insights into hardware behavior.System Event Monitoring: Monitor critical system events such as context switches and page faults.Top-Down Microarchitecture Analysis (TMA): Master this technique to write efficient code and maximize CPU performance.Flame Graphs: Learn how to create and interpret flame graphs to pinpoint performance bottlenecks.Linux Tracing Frameworks: Explore tracing tools like tracepoints, kprobes, and uprobes to capture in-depth performance data.Kernel Function Tracking: Gain insights into kernel activities and identify points of failure like segmentation faults.Application Function Tracing: Track application behavior to diagnose and resolve performance issues.Custom Performance Analysis: Learn to design custom analyses, such as calculating function latency.Hands-On Examples: Practice with real-world examples to build practical performance engineering skills.By the end of this course, you'll have the confidence to identify, diagnose, and optimize performance issues in your applications and systems as part of your daily work.
