This article consolidates practical knowledge on system performance optimization, covering key metrics, load‑testing utilities, Linux monitoring commands, and JVM profiling tricks to help engineers pinpoint and resolve throughput, latency, CPU, disk, and network bottlenecks.
This article explains how to employ the open‑source ApacheBench (ab) tool to conduct load‑testing and dynamic monitoring of a Linux‑based load‑balancing module, detailing environment setup, command parameters, and result analysis for effective performance evaluation.
This article provides a comprehensive overview of performance testing, explaining its definition, key metrics such as throughput, response time and concurrent users, the relationship between these indicators, optimal testing timing, a step‑by‑step workflow from requirement analysis to reporting, and practical tips for effective load testing and system optimization.
This article explains why rate limiting is essential for flash‑sale (seckill) services, compares token‑bucket and leaky‑bucket algorithms, and provides practical configuration examples for Tomcat, Nginx, and OpenResty, along with testing methods and code snippets.
The article describes a production traffic replay system that records real user traffic, creates scalable pressure sources, supports both 4‑layer and 7‑layer protocols, and provides automated fail‑over and monitoring features to enable realistic functional and performance testing at large scale.
After a recent feature release caused a sharp decline in a key data metric, the team followed a systematic, fourteen‑step troubleshooting process—including verification, code review, DBA involvement, local debugging, environment comparison, logging, packet capture, service restarts, request mode changes, load testing, and partition resizing—to identify and resolve a Kafka‑related throughput bottleneck.
This article presents a comprehensive case study of Baidu Nuomi's online load‑testing methodology for high‑traffic marketing events, covering capacity estimation, test planning, execution, anti‑attack measures, platform architecture, and lessons learned to ensure system reliability and performance under peak loads.
This article explains the design of an extended NSQ message format, illustrates its application in load‑testing, service isolation, and channel‑level filtering scenarios, describes the NSQ migration proxy tool and Spark/Flume connectors, and outlines future enhancements such as flow control and batch subscription.
This article explains JD Finance's operational challenges in a rapidly expanding micro‑service environment and presents a comprehensive approach that combines offline and online load testing, precise capacity calculations, and intelligent root‑cause alert analysis using both rule‑based and machine‑learning techniques.
This article details how Xiaohongshu tackled rapid growth challenges by designing, implementing, and operating a full‑link performance testing platform in just two days, covering system architecture, testing models, collaborative deployment, capacity planning, and practical advice for teams seeking reliable e‑commerce load testing.
This article explains how to improve Java application performance by defining measurable metrics, using load‑testing tools like Gatling, monitoring with APM solutions such as Stackify Retrace, applying code‑level best practices, tuning JVM heap and garbage collection, optimizing JDBC access, and scaling the architecture.
This article explains the purpose of performance (stress) testing, compares four testing approaches, details the chosen proportional‑deployment strategy, and provides comprehensive preparation steps, script guidelines, metric analysis, and practical tips for ensuring system stability during large‑scale traffic spikes.
This article explains how Dada's order platform was refactored from a Python Flask service to a Java‑based microservice architecture, employing load testing, circuit breaking with Hystrix, and multi‑datacenter disaster recovery to reliably handle eight million orders in a single day during the Double‑11 shopping festival.
This article explains how to measure a system's load capacity, identifies key factors such as bandwidth, hardware, OS and application configurations, and provides practical optimization techniques for Linux, Tomcat, Nginx, MySQL, and Redis to handle high concurrency efficiently.
A detailed case study reveals how an out‑of‑memory (OOM) situation on a Linux server caused Tomcat to crash during load testing, explains the kernel's OOM killer behavior, and provides practical solutions such as enabling swap and tuning Apache limits.
This article examines JD.com’s 618 shopping festival performance, the deployment of unmanned delivery robots, and the design and architecture of the ForceBot full‑link load‑testing system that enabled precise capacity planning and bottleneck detection for massive e‑commerce traffic.
This article explains why load testing is essential, compares three popular tools—ab, wrk, and Locust—including their command‑line usage and key parameters, and offers practical tips and additional options for accurate performance testing.
This article explains Alibaba's four‑step capacity‑planning methodology, the various single‑machine load‑testing techniques, the design of a full‑link load‑testing platform for Double‑11, and the dynamic flow‑control framework that together ensure system stability during extreme traffic spikes.
This article explains how to design performance testing scenarios, covering test models, metrics, script preparation, concurrency calculations, pressure strategies, run times, delay settings, user termination, monitoring methods, and various typical scenario types such as baseline, load, mixed, capacity, large‑concurrency, stability and scalability tests.
This guide introduces St-load, a Linux‑based streaming load testing tool, and provides step‑by‑step instructions for installing, compiling, and using its RTMP, HTTP, and HLS testing features, including command‑line examples and parameter explanations for both push and pull scenarios.
This article demonstrates how to install and run load‑testing tools Artillery and Apache ab, explains their command‑line options, interprets the performance metrics they produce, and provides tips for handling system limits during high‑concurrency tests.
ForceBot is JD.com’s comprehensive full‑chain load‑testing platform that simulates user behavior across the entire purchase flow, isolates test traffic, leverages Docker‑based agents, GRPC services, and real‑time data analytics to identify bottlenecks, optimize resource planning, and support both routine and peak‑traffic scenarios.
This article recounts Alibaba's 7‑year journey of capacity planning for the massive Double 11 shopping festival, detailing early guesswork, the introduction of load‑testing, online and scenario‑based testing, traffic isolation, and full automation that enabled precise resource allocation across hundreds of services.
This article revisits the technical design and evolution of the 2015 Chinese New Year Gala "shake" activity, detailing how the backend architecture was progressively refined—from a simple prototype to a production‑grade system capable of handling tens of millions of concurrent requests—through resource pre‑download, access‑layer integration, load‑balancing, and robust fail‑over mechanisms.
This guide introduces the wrk HTTP load‑testing tool, explains its multithreaded and Lua‑scriptable architecture, walks through installation and command‑line usage, interprets key performance metrics, and demonstrates how to extend functionality with custom Lua scripts for advanced testing scenarios.
This report details the performance testing of the 360 Endpoint Security Management System, covering product overview, architecture, tool selection (favoring LoadRunner), test scenarios, script implementation, execution results, resource utilization, and analysis, concluding with insights on testing methodology and optimization.
This guide explains how to design and test high‑concurrency systems by choosing appropriate server architectures, load‑balancing, database clustering, caching layers, message queues, first‑level caches, and static data strategies to ensure smooth operation under heavy user traffic.
This article compares JMeter, LoadRunner, Locust, asyncio (Python) and Go-based load testing tools by measuring CPU, memory, Nginx response and TPS under varying concurrency levels, concluding that Go delivers the highest throughput while Python coroutines offer significant improvements over older Python versions.
This article details Meituan's technical approach to handling massive traffic spikes during large‑scale promotions, covering background of the O2O deal platform, traffic‑model construction, capacity‑budget calculations, micro‑service architecture evolution, pressure‑test strategies, and the PTP performance‑testing environment used to validate system limits.
This guide explains what load testing is, how to use ApacheBench (ab) with simple commands, interprets the key performance metrics it returns, and highlights important precautions for effective and safe web server stress testing.
This guide explains how to use GoReplay to capture live HTTP requests from a production environment, filter and store them, and replay them in a test environment with adjustable speed, enabling realistic load testing and debugging without affecting real users.
Benchttp is a minimal‑weight Go‑based load‑testing utility that lets you measure QPS, duration, and concurrency with simple command‑line flags, and the article explains its usage, common parameters, and step‑by‑step installation on Linux.
With increasing emphasis on user experience, this article introduces five widely used web performance testing tools—HP LoadRunner, Apache JMeter, Load UI, Httperf, and Siege—detailing their capabilities for load simulation, functional testing, and performance analysis to help developers and testers optimize application reliability and speed.
This article explains which external and internal performance metrics matter for backend services, how to interpret them, common bottlenecks, and provides a step‑by‑step example using JMeter to measure throughput, latency, CPU, memory, load, network and disk I/O.
This article explains how to assess the maximum concurrent user capacity of short‑connection server implementations by using the HTTP Transaction Rate metric, contrasting it with long‑connection testing, and provides practical calculation examples and terminology definitions.
This guide explains how tcpcopy copies live traffic to test machines, details its installation, demonstrates performance testing commands, and presents results and monitoring graphs to show the impact of replicated traffic on backend and test servers.
This article presents a detailed comparison of six server‑side load‑testing tools—LoadRunner, JMeter, Spirent Avalanche, Siege, Tsung, and Locust—through multiple real‑world scenarios, evaluating their protocol support, scripting flexibility, performance, cost, and suitability for different testing requirements.
This article explains a systematic, waiting‑point‑based approach to enterprise Java performance tuning, covering load‑test design, analysis of existing versus new applications, hierarchical and technical waiting points, pool and cache sizing, and a back‑tuning workflow to achieve measurable improvements.
This article outlines a complete performance‑testing workflow—including requirement analysis, overall and detailed design, environment selection, execution, result collection, and bottleneck analysis—using a wallet "red‑packet" project that must handle tens of thousands of requests per second.
This article outlines ten practical guidelines—ranging from test environment consistency and dedicated hardware to network capacity checks, OS tuning, realistic workloads, proper test duration, and comprehensive result reporting—to ensure reliable and reproducible HTTP server performance benchmarks.
This guide explains key metrics, influencing factors, and practical tuning steps—including bandwidth, hardware, OS limits, TCP parameters, and server configurations—to assess and improve a backend system's maximum request handling capacity under high concurrency.
Suning’s technical team shares how they prepared for Double‑11 by splitting monolithic services into focused modules, building a robust foundational platform with cloud, middleware, and monitoring tools, refining R&D processes, and implementing comprehensive load‑testing, optimization, and emergency response plans to ensure system stability under massive traffic.
This article explains essential software performance testing concepts—including QPS/TPS, concurrency, response time, throughput calculations, think time, and how to estimate daily traffic using real‑world examples from e‑commerce platforms—providing formulas and practical guidance for engineers, administrators, and developers to assess and optimize system capacity.
This article introduces load‑generating tools, compares their performance, and provides a step‑by‑step Tsung installation and XML configuration to help engineers build a web cluster capable of handling millions of requests per second.
This article explains how to achieve ultra‑high HTTP request rates by using load‑generating tools such as Apache Bench, httperf, JMeter and Tsung, discusses Linux socket limits, provides step‑by‑step installation and configuration examples for Tsung on CentOS, and shows how to run tests and interpret the results for building a high‑performance web cluster.
This guide walks you through installing and configuring Nginx for extreme request rates, optimizing kernel TCP settings, and using Tsung to benchmark a server that can sustain more than 900,000 requests per second, with detailed command examples and configuration snippets.
This article compiles a comprehensive discussion of load‑testing tools, profiling techniques, and optimization strategies for web and backend services, offering practical advice on selecting utilities, analyzing performance bottlenecks, and automating testing workflows.
Learn about ten free, open‑source tools—including Grinder, JMeter, and Siege—that let you simulate concurrent users, measure performance, and identify bottlenecks in web applications, helping you assess server capacity and optimize site responsiveness under load.
Tcpcopy is a distributed online pressure testing tool that copies live traffic to test machines, enabling realistic simulation of production environments without deploying new code, and the article explains its principles, code details, deployment steps, advanced usage, and real‑world application at Yitao.