This article explains how to estimate and design system capacity—including traffic, concurrency, QPS, and resource requirements—by illustrating concepts, analytical formulas, peak‑load assessment techniques, performance testing, and a practical case study of a book‑reservation system.
Jeff Geerling compared the time it takes to compile a Raspberry Pi OS Linux kernel on an Intel i9‑powered MacBook Pro, an M1 Mac mini and an M1 MacBook Air, finding the M1 machines finish up to 30% faster while staying cooler and quieter.
This article details Dada Group's full‑link load‑testing architecture, from industry background and machine‑tagging design to platform components, deployment workflow, optimization plans, fine‑grained modeling, pre‑heat strategies, post‑test analysis, and the resulting stability and efficiency gains.
This article explains how to design and implement distributed performance testing cases using jar‑method invocations, demonstrates a Java demo with reflection‑based execution, discusses common pitfalls, and shows how to package, deploy, and run the test cases within a Docker‑enabled environment.
The JD Health Technical Product Team organized a comprehensive 618 promotion preparation in 2021, covering system architecture reviews, capacity assessments, performance stress testing, offensive‑defensive drills, system optimization, 24‑hour monitoring, and product operation training to ensure high availability and stable service during the large‑scale sales event.
This article details how Lakala built a distributed, cloud‑native messaging platform using Apache Pulsar, covering functional requirements, architectural advantages, performance testing, and real‑world integration scenarios such as OGG adapters, TiDB pipelines, OpenMessaging, custom sources, functions, Flink connectors, and future plans.
This article shares three years of experience building an automated performance‑testing platform for Alibaba Cloud Elastic Compute, covering what performance testing is, how to design comprehensive tests for CPU, memory, storage, network and OS, and the essential skills engineers need to become true performance architects.
This article introduces the Python‑based Locust load‑testing framework, explains its event‑driven architecture, weight‑based task distribution, command‑line and web UI operation, distributed execution, and practical usage patterns, then evaluates its advantages, drawbacks, and suitability for performance testing projects.
This article details the billing group's challenges with single‑node MySQL, the evaluation of alternative databases such as TiDB and CockroachDB, performance testing, migration strategies, operational best practices, and the final decision to adopt TiDB for scalable, high‑availability data services.
This article explains the purpose of the TCP TIME_WAIT state, how it prevents packet loss and connection errors, examines its duration based on the Maximum Segment Lifetime, and analyzes why high‑QPS load testing tools like ab may appear to bypass TIME_WAIT, revealing the impact on server resources.
This guide explains how to extend the moco API with request‑hit monitoring, demonstrates building a lightweight test service, and shows achieving over 50,000 QPS using FunTester, while detailing the available assertion methods and their practical code examples.
Over four years at Ele.me, the author chronicles the transformation of the order platform from a single‑machine Python monolith to a distributed, sharded, message‑driven architecture, detailing challenges in scaling, performance testing, fault injection, Redis usage, and service separation that shaped today’s robust backend.
This article explores how to accurately measure QPS in Java performance‑testing scripts by using JsonPath and regular‑expression validation, analyzes error margins under different thread and iteration configurations, and demonstrates that exception handling has minimal impact on overall throughput.
This document outlines a detailed testing strategy for the Cloud Classroom mobile app, covering security, static code analysis, performance, compatibility, and overall quality assurance to ensure stable functionality and a positive user experience.
Facing rising traffic, JD Daojia’s in‑house load‑testing platform was redesigned to automate script management, enable distributed JMeter execution, integrate real‑time monitoring, and support custom RPC protocols, dramatically lowering manual effort, cutting test cycles from an hour to fifteen minutes while ensuring system stability.
This article examines how adding Java thread‑safe classes and the synchronized keyword influences QPS in performance testing, presenting code modifications, benchmark results for 20 and 40 threads, and concluding that their impact on measured throughput is minimal.
This article introduces Locust, an open‑source Python load‑testing framework, explains how to install it (including optional pyzmq for distributed mode), describes its core components, walks through a sample test script with database integration, and shows how to run and interpret results via the web UI or command line.
The article explains that mysqlslap appears to produce smoother latency results than sysbench because mysqlslap reports metrics per test round rather than per individual SQL statement, leading to mis‑interpretation of Max/Avg/Min values, while sysbench can reveal per‑statement latency variations.
This article explains how to work around Groovy's invokeMethod limitation of accepting only a single Object argument by using String arrays or var‑args, demonstrates Java and Groovy reflection implementations, and provides complete code examples for multi‑parameter performance test scripts.
The guide explains YouZan’s low‑code BOS platform, its component‑based architecture, and a multi‑level testing methodology—including automated component, product, and workflow tests, traffic replay, performance checks, and a robust fallback mechanism—to ensure seamless migration and reliable operation of digitized business processes.
This article experimentally examines how different sources of performance‑test error—fixed versus random request latency, thread‑count variations, and log‑printing overhead—affect QPS calculations, revealing that higher concurrency amplifies error while increased request repetitions and reduced logging improve measurement accuracy.
This article explains the concept of soft‑start in performance testing, presents Java implementations for both fixed‑thread and fixed‑QPS models, discusses error‑impact considerations, and provides practical code snippets to gradually ramp up load and improve measurement accuracy for high‑throughput services.
The article chronicles Eleme’s four‑year journey of transforming its order platform—from the early Python‑based Zeus monolith through team formation, service decoupling, sharding, message broadcasting, testing automation, Redis and cache redesign, to virtual product handling and governance—highlighting practical challenges, decisions, and hard‑won engineering insights.
This article presents a systematic analysis of how thread numbers, request counts, response times, and response‑time variance affect performance‑testing error rates, using a Java demo that simulates single‑API calls and reports detailed JSON metrics for each configuration.
This article explains how to extend a performance testing framework with new QPS and deviation fields, implement accurate QPS calculations for both fixed‑thread and fixed‑QPS models, and prepare the system for quantitative error analysis using detailed Java code examples.
This article demonstrates how a MySQL DBA can use the open‑source analysis_binlog tool to efficiently parse large binlog files, count DML operations, convert binlog to SQL, and perform parallel analysis after setting up a CentOS test environment and running sysbench stress tests.
This article examines the challenges of chain performance testing without branch handling, proposes logical branch controls and synchronized test termination, and provides concrete Java code examples to improve reliability and accuracy of load‑testing pipelines.
The article shares a senior developer’s firsthand observations on how architects differ in attitude, technical depth, system‑level responsibilities, and collaboration, offering practical steps and mindset changes needed to transition from senior engineering to a full‑stack architecture role.
This article explains how to enrich a performance‑testing framework with an asynchronous progress bar that also reports real‑time QPS, detailing the design of a unified Progress class, its handling of different thread models, the QPS calculation logic, a sample Groovy test script, and the resulting console output.
This article explains how to extend a performance testing framework to record and visualize individual HTTP interface latencies by parsing consolidated logs, grouping timings by URL, and using a Groovy script with StatisticsUtil to generate response‑time charts.
This article explains how to conduct stress testing using JMeter, monitor Java applications with jconsole and jvisualvm, optimize gateway and microservice performance, improve database query speed with indexing, and tune garbage collection by enlarging the Eden space, providing step‑by‑step instructions and results.
This article explains what performance testing is, introduces the four core system metrics—response time, throughput, resource utilization, and concurrency—illustrates their relationships, details how request latency is composed, and presents practical tuning techniques and real‑world case studies to improve system performance.
This article explains what a prefix (partial) index is in MySQL, why it can reduce index size and improve query speed, outlines suitable scenarios and limitations, and demonstrates performance differences through a series of tests with varying prefix lengths on a large test table.
This article explains how to eliminate the serial‑login bottleneck in multi‑user performance testing by using Java’s CyclicBarrier to synchronize user logins, then runs a two‑step insert‑update scenario with a custom FunTester class, complete with full code examples and result analysis.
This article explains how to transform a Java HttpRequestBase object into an equivalent curl command, enabling rapid request replay across servers and detailed timing analysis, and demonstrates the process with step‑by‑step code examples and a functional test using the FunTester framework.
This article explains practical techniques for generating varied request parameters—such as random numbers, thread‑safe identifiers, random strings, and upstream data extraction—to avoid identical inputs in multi‑call performance tests and improve test reliability.
iQIYI built a Netty‑based WebSocket long‑connection gateway that centralizes session management, uses RocketMQ for cluster‑wide broadcast, provides a simple HTTP push API, scales horizontally to millions of connections, integrates Prometheus/Grafana monitoring, and dramatically cuts development effort for real‑time features such as comments and authentication.
This article presents a systematic experiment that compares four different IntelliJ IDEA memory configurations on a MacBook Pro, measuring IDE startup time, project loading speed, and JVM garbage‑collection metrics with jstat to identify the most efficient setup for large Java monoliths and micro‑service projects.
This article walks through the design, scenario planning, and Java implementation of a fixed‑thread load test that simulates teacher login, knowledge‑point queries, course recommendations, and collect/uncollect actions, then presents the resulting performance metrics.
This article walks through the design and implementation of a Java RPC framework, detailing added serialization options, multiple load‑balancing strategies, client‑side service caching, TCP long‑connection handling with Netty, performance testing, and provides the full source code for replication.
This article translates and expands on a HikariCP wiki post, explaining why smaller database connection pools often yield better performance, presenting benchmark videos and data, describing the underlying CPU, disk, and network constraints, and offering a simple formula to calculate an optimal pool size for typical server configurations.
This article introduces the fundamentals of full‑chain load testing, explains why it is essential for large‑scale distributed systems, outlines the DESP model with its four simulation dimensions, and presents a detailed case study of New Oriental's continuation‑class platform including architecture, data preparation, load design, automation and recruitment information.
This guide explains how to use MySQL’s built‑in mysqlslap tool and the third‑party sysbench utility to perform comprehensive database stress testing, covering configuration changes, command examples, result interpretation, and additional CPU and I/O benchmarks for accurate performance evaluation.
During the Chinese New Year break the author used draw.io (now diagrams.net) to create a comprehensive architecture diagram of the FunTester testing framework, outlining its functional and performance layers, supported protocols, utility modules, and multithreaded execution model.
This article presents a practical demo for multi‑chain performance testing with FunTester, describing the overall workflow, custom multithreaded implementation, Java code example, and sample console output to illustrate how to execute and analyze load tests across multiple request paths.
This guide explains how to configure MySQL for high‑concurrency testing, use the built‑in mysqlslap tool and the third‑party sysbench utility to benchmark read/write performance, CPU, I/O, and memory, and interpret the resulting metrics for database optimization.
This article presents a seven‑day experiment measuring how much memory a TCP connection occupies in the kernel under various states such as ESTABLISH, data transfer, and TIME_WAIT, and summarizes the kernel's memory‑optimisation techniques for sockets.
This article explains how to capture all HTTP requests from a browser, filter out irrelevant assets, convert them into HttpRequestBase objects, and run a concurrent load test across multiple APIs using the FunTester framework, complete with code snippets and performance metrics.
This article explains how to capture browser curl requests, parse them into Java HttpRequestBase objects, and run high‑concurrency performance tests using a custom FunTester framework, complete with code examples, metrics, and repository links.
TXRocks is a transactional storage engine based on RocksDB that offers significant space savings compared to InnoDB while maintaining comparable performance, with optimizations including sum operator pushdown and AEP-based secondary caching to address RocksDB's read performance limitations.
The article introduces Baidu's self‑developed test middleware, explains its data‑plane and control‑plane architecture inspired by Istio, describes the challenges of building test environments, and details the system’s components, working modes, and diverse testing use cases such as functional, system, integration, sandbox, and chaos engineering.
This article details a systematic full‑link performance testing workflow—including background, timing, scenario design, data preparation, capacity planning, monitoring, issue analysis, and post‑test cleanup—aimed at reliably evaluating and scaling e‑commerce services during major promotional events.
This article clarifies the distinction between stress testing and load testing, describing their respective models—fixed‑thread and fixed‑QPS—along with key metrics, formulas, and practical benefits for evaluating system performance under varying workloads.
This article examines MySQL's recommendation to use auto_increment primary keys instead of UUID or random snowflake IDs by creating three tables, running insertion benchmarks with Spring Boot and JdbcTemplate, and analyzing how index structure affects performance and scalability.
This article introduces JMH, the Java Microbenchmark Harness, explains why traditional main‑method benchmarks are unreliable, and provides step‑by‑step instructions, code examples, and best‑practice annotations for accurately measuring method‑level performance in Java applications.
This article details a systematic experiment that compares default, modest, balanced, and aggressive JVM memory configurations for IntelliJ IDEA, measuring startup time, project load speed, and garbage‑collection behavior on a large monolith and micro‑service projects, and concludes with practical tuning recommendations.
This article investigates how different IntelliJ IDEA memory configurations affect IDE speed and responsiveness by testing default, large, balanced, and sophisticated settings on a MacBook Pro with a monolithic Java project and microservices, measuring startup times, project loading, and JVM garbage collection using jstat.
This article explains how to use Java's CyclicBarrier and related concurrency utilities to parallelize user login preparation for load testing, reducing setup time and synchronizing multi‑stage performance tests with concrete code examples and sample console output.
This article examines common misconceptions about Java concurrency utilities such as ThreadLocal, ConcurrentHashMap, and CopyOnWriteArrayList, demonstrates real‑world bugs caused by thread reuse and non‑atomic operations, and provides concrete solutions and performance‑tested alternatives.
This article explains how thread reuse in a Tomcat thread pool can cause user‑information leakage when ThreadLocal is misused, analyzes the non‑atomic behavior of ConcurrentHashMap operations, demonstrates performance differences between locking and atomic methods, and warns against inappropriate use of CopyOnWriteArrayList in high‑write scenarios.
The presentation outlines the AI engineering efficiency platform covering algorithm metric and evaluation, micro‑service performance testing, and dataset management architectures, detailing business pain points, platform‑wide improvements, technical designs, real‑world demos, and future directions to achieve accurate, fast, and stable AI services.
This article presents a performance comparison of Dubbo 2, Dubbo 3 interface‑level, and Dubbo 3 application‑level address discovery models under a million‑instance scenario, showing significant reductions in memory usage and GC frequency for the newer models.
To overcome HBase’s weak availability and GC‑induced latency spikes, the DiDi team introduced a replication‑based client multi‑read (hedged‑read) mechanism and migrated to the Z Garbage Collector, which together dramatically cut maximum and 99.9th‑percentile latencies while keeping services online during region disruptions.
The article shares the author’s personal approach to strengthening technical abilities by combining deep theoretical study with hands‑on experimentation, recounts real‑world problem‑solving experiences, and recommends a curated list of classic books and resources for developers seeking solid fundamentals.
This guide outlines the three‑stage data‑migration testing process—preparation, execution, and verification—detailing how to assess data scope, choose migration methods, evaluate system compatibility, construct test scenarios, validate scripts, monitor performance, and confirm backend, frontend, and business‑logic integrity after migration.
This article introduces a reusable Java WebSocket client wrapper that adds convenient date handling, message logging, configurable constants, and performance‑testing utilities, explains its implementation details with full source code, and provides a simple demo showing connection, messaging, and graceful shutdown.
This article details how a performance testing team at a major e‑commerce company designed and evolved full‑link load testing from zero to a mature, repeatable process, covering its definition, historical milestones, challenges during major sales events, and future optimization plans.
This article explains how to employ JMeter's pre‑processor (HTML Link Parser) and post‑processor (Result Status Action) to create dynamic URL requests, simulate concurrent users, handle error responses, and visualize results for performance testing of web services.
The article examines MySQL pagination performance on a large order_history table, presenting various pagination methods—including simple LIMIT, subquery, ID range, and temporary table techniques—along with detailed test results that show how query speed varies with record count and offset.
Using sysbench, this study benchmarks MySQL 5.7.22 and MySQL 8.0.15 under read‑write, read‑only, and write‑only workloads with various sync_binlog and innodb_flush_log_at_trx_commit settings, revealing performance trends across different concurrency levels and highlighting that MySQL 5.7 often outperforms 8.0 in these tests.
This article presents a systematic performance comparison between Java and Go microservices—covering raw speed, memory usage, logging impact, GraalVM native images, and Kubernetes deployments—to reveal when each language excels and the practical trade‑offs involved.
This article explores four practical methods—batch inserts, transaction wrapping, ordered primary‑key insertion, and their combined use—to dramatically improve MySQL InnoDB insert performance, presents test results across various data volumes, and offers key configuration cautions.
This article explains how to design and implement a bespoke performance testing platform for web pages using Lighthouse, Puppeteer, and Node.js, covering architecture, data collection, hook functions, gatherer modules, scoring logic, and automated weekly reporting.
This article revisits a previous logback configuration test, expands the sample size and repeats the measurements using JMeter, JProfile, and New Relic APM to compare synchronous and asynchronous logging in Logback and Log4j2, revealing that asynchronous logging reduces log latency but only marginally improves overall TPS, while Log4j2’s async implementation can boost TPS up to six times.
This guide introduces thirteen practical Linux utilities—ranging from network bandwidth monitors and disk I/O testers to security hardening and terminal multiplexing tools—complete with installation commands, usage examples, and key options to help system administrators efficiently monitor, diagnose, and optimize their servers.
This guide introduces thirteen practical Linux utilities—including Nethogs, IOZone, IOTop, IPtraf, IFTop, HTop, NMON, MultiTail, Fail2ban, Tmux, Agedu, NMap, and Httperf—covering their purpose, installation commands, key options, and example usage for effective system monitoring and troubleshooting.
This article examines the growing need for comprehensive IoT testing, outlines how to select representative devices, and details the four main testing domains—security, performance, functionality, and compatibility—while offering practical guidance for QA teams to ensure high‑quality connected products.
This guide introduces thirteen practical Linux operation tools—including Nethogs, IOZone, IOTop, IPtraf, IFTop, HTop, NMON, MultiTail, Fail2ban, Tmux, Agedu, NMap, and Httperf—detailing their purpose, installation commands, usage examples, and download sources to help administrators efficiently monitor and manage system resources.
This article examines various MySQL pagination strategies—including simple LIMIT clauses, sub‑query optimizations, ID‑range queries, and temporary‑table methods—by testing them on a table with millions of rows and presenting detailed timing results that reveal significant speed improvements.
This article examines Eastern Securities' transition to a microservice architecture by detailing the design, implementation, and performance of its gRPC‑Nebula service‑governance framework and the Star‑Chen platform, covering challenges such as heterogeneous interfaces, service registration, load balancing, fault tolerance, traffic control, multi‑registry support, and real‑world deployment results.
This article walks through the design, parameterization, and Java implementation of load‑testing scripts for two related APIs—one that creates a resource and another that deletes it—covering special handling of duplicate IDs, method encapsulation with default arguments, concurrent execution, and visual result analysis.
This article details a performance comparison between UCloud's KuaiJie‑based UDB service and self‑built MySQL instances, covering environment setup, sysbench benchmarking, result analysis, and cost evaluation to guide optimal database instance selection.
This article presents a systematic performance test of Hibernate and MyBatis covering bulk inserts, single‑table queries, and various association queries on MySQL, analyzes the impact of lazy versus eager loading, caching, and data volume, and provides practical recommendations for choosing the appropriate ORM framework.
ESrally is the official Elasticsearch benchmarking tool; this guide walks through its installation prerequisites, step‑by‑step setup of Python, JDK, and Git, configuration of tracks, cars, pipelines, and challenges, and demonstrates real‑world performance comparisons across Elasticsearch versions and hardware platforms.
JMeter’s reported throughput can be misleading because it includes local processing time, especially when response validation such as regex extraction adds overhead, leading to significant underestimation of actual server load; the article demonstrates this with experiments and suggests micro‑benchmark corrections to obtain accurate results.
This article introduces MySQL 8.0.20 Group Replication, covering single‑master and multi‑master modes, monitoring, failover procedures, abnormal recovery, flow control, performance testing, encountered issues, and limitations, and provides a downloadable PDF with detailed documentation hosted on Baidu Cloud.
Through systematic performance testing, caching strategy analysis, server configuration tuning, Elasticsearch shard optimization, and validator refactoring, the team transformed the recommendation service’s throughput from roughly 190 QPS to over 2,200 QPS, demonstrating how targeted backend optimizations can dramatically improve high‑concurrency applications.
This case study details how the Tangdou square‑dance app leveraged big‑data analytics, cloud deployment, and performance testing to solve user‑acquisition, crash‑handling, and operational challenges, ultimately achieving rapid growth and a robust, scalable data‑driven platform.
Through extensive JMeter and APM testing on a 6‑core, 8 GB server, this article reveals that switching Logback from synchronous to asynchronous logging yields minimal TPS improvement, explains why, and demonstrates that Log4j2’s async implementation can achieve up to six‑fold performance gains.
This article explains why large‑scale Elasticsearch deployments need rigorous performance testing, compares available testing tools, walks through installing and configuring the official Rally benchmark suite, details hardware recommendations, shows how to run tests against multiple cloud providers, and teaches you how to interpret the resulting metrics to make informed cluster‑selection decisions.
The article explains how QA engineers should adapt their testing approach when a live‑streaming service evolves from a simple business‑focused system to a modular, scalable platform, covering architecture changes, testing directions such as interface, data preparation, performance, and comprehensive project control.
This article details how the Alibaba Nacos service registry was evaluated and chosen over Eureka, Zookeeper, and Consul for a large‑scale micro‑service platform, presenting comprehensive performance benchmarks, functional API tests, sync behavior analysis, and automated JUnit testing with real‑world deployment data.
This article explains practical techniques—equivalence partitioning, boundary value analysis, and scenario testing—to create thorough login test cases, then expands with advanced functional, security, performance, and compatibility scenarios, helping testers improve coverage and demonstrate expertise in interviews.
This article details Beike's journey of storing and querying a 480‑billion‑triple industry graph in milliseconds, covering graph database fundamentals, a comparative evaluation of JanusGraph and Dgraph, the design and deployment of a Docker‑K8s based Dgraph platform, data ingestion pipelines, a custom Graph‑SQL layer, performance testing, optimizations, and future roadmap.
This guide explains how to create a globally unique, auto‑incrementing variable in JMeter by using a synchronized Groovy script that leverages the built‑in props object, demonstrates setting an initial value, and compares locked versus unlocked execution results.
The author shares a series of JMeter demos that leverage Groovy for context access, variable handling, command execution, request manipulation, regex extraction, cookie and header processing, and throughput error analysis, while recommending thorough documentation review and community interaction to deepen JMeter expertise.
The article explains how to isolate and measure the time spent on request signature generation in an API performance test using micro‑benchmarks, provides Java and Groovy code examples, and discusses subtracting non‑request overhead to obtain calibrated throughput.
Alibaba Entertainment transformed its capacity management from manual, experience‑based decisions to a fully automated system that continuously evaluates single‑machine performance, identifies performance and success‑rate breakpoints, and drives elastic scaling, dramatically improving resource utilization, availability, and development efficiency across all its applications.
API testing, a core form of backend functional testing, is essential because modern services share a single set of interfaces across web, mobile and mini‑programs, and mastering it provides the foundation for automation and performance testing while boosting employability.
This article explains MySQL benchmark testing, the role and metrics of benchmarking, introduces sysbench, provides step‑by‑step installation and usage instructions, and offers practical tips for interpreting results and optimizing database performance in production environments.
