This article presents a practical guide for SpringBoot projects, detailing twelve concrete strategies—including batch processing, asynchronous execution, caching, pagination, and transaction handling—to dramatically reduce API latency and improve overall backend efficiency.
Learn practical PHP strategies to lower server load—including using Memcached or Redis for data caching, implementing pagination to limit query size, and caching computational results—complete with clear code snippets and guidance on adapting these techniques to improve performance and stability.
This article explains the unique challenges of flash‑sale systems—massive, short‑lived traffic spikes—and presents practical backend optimizations such as front‑end request filtering, Redis‑based atomic counters, streaming queues, safe database updates, and unit‑level isolation for global deployments.
This article explains the concept of Nginx static‑dynamic separation, describes three deployment patterns—mixed, fully separated, and CDN‑backed architectures—illustrates how static assets are served directly by Nginx to reduce backend load, and highlights performance benefits such as lower latency, higher throughput, and easier scaling.
This guide explains how to use Alibaba's open‑source Java diagnostic tool Arthas to non‑intrusively monitor, locate, and resolve CPU performance bottlenecks in production Java applications, covering installation, key commands, a step‑by‑step troubleshooting workflow, and a real‑world case study.
This article explains how merging similar or duplicate requests upstream with Hystrix Collapser, a custom BatchCollapser, and Guava's ConcurrentHashMultiset can dramatically reduce downstream load, improve system throughput, and outlines their implementations, configurations, and ideal use‑cases.
This article examines the severe row‑lock competition in MySQL during hot‑SKU inventory positioning, analyzes its causes and risks, and presents a comprehensive set of mitigation strategies—including DB account isolation, flow‑control, request merging, and message‑queue optimizations—validated by extensive performance testing and successful rollout across multiple clusters.
Bilibili’s ticketing platform evolved from a single‑transaction, synchronous design to an asynchronous batch system and finally to a Redis‑cached inventory layer, adding DB isolation changes, sharding, bloom‑filter protection and adaptive rate limiting, which together enabled handling up to 930 k requests per second and stable high‑concurrency ticket sales.
This article walks through diagnosing slow API responses, using Pfinder for tracing, refactoring Java collection handling with Sets, and applying JSF asynchronous calls to dramatically cut latency, offering a step‑by‑step guide to backend performance optimization.
This article explains how Nginx uses an event‑driven, asynchronous non‑blocking I/O model together with epoll/kqueue and various optimization strategies such as keep‑alive, caching, efficient data structures and load balancing to handle millions of concurrent connections in large‑scale internet architectures.
This article compares three approaches for verifying username uniqueness during app registration—direct database queries, Redis caching, and Redis-backed Bloom filters—detailing their performance, scalability, memory usage, and trade‑offs, and provides Java code examples for each solution.
Facing tight schedules and diverse coding habits, many Chinese projects struggle with API latency; this guide walks through requirement analysis, acceptance standards, common pitfalls, and a comprehensive set of optimization tactics—from configuration and code tweaks to caching, async processing, and observability tools—to dramatically improve API performance.
This comprehensive guide explains how to identify and resolve slow web page performance by addressing hardware resources, network configuration, and both front‑end and back‑end optimizations such as server upgrades, CDN usage, HTTP/2, image compression, code minification, caching, and efficient database queries.
This comprehensive guide explains hardware, frontend, and backend optimization techniques—including server upgrades, CDN usage, code minification, caching strategies, and database tuning—to systematically improve website speed and maintain high performance.
This article analyzes why failed file downloads often display a blank JSON page and compares two technical solutions—a pre‑check API and an API that returns a download URL—evaluating their latency, security, user experience, and server‑resource impacts before recommending the optimal approach.
This article presents a detailed three‑step plan used by Zhezhuan to detect and eliminate zombie services, unused code methods, and obsolete component dependencies through monitoring, static analysis with Spoon, and Java‑agent based runtime tracing, achieving significant resource savings and improved code health.
This article presents Qunar Travel's "system slimming" project, describing how observability techniques, a two‑stage strategy, and automated tooling were used to identify and remove unused services and code, achieving a 50% reduction in code size, a 26% cut in services, and measurable improvements in reliability and release efficiency.
By instrumenting Java services with JaCoCo’s agent and analyzing runtime coverage data, teams can identify unused (“zombie”) code, safely remove or deactivate it, and consequently reduce maintenance costs, shorten delivery cycles, and improve overall system performance and developer productivity.
This article explains how to use a Bloom filter backed by Redis bitmap and Roaring Bitmap sharding to efficiently filter 800 million OpenID queries, covering memory planning, hash function selection, code implementation, and performance‑tuned batch write strategies.
This article explains how to merge multiple backend requests in a SpringBoot application using a blocking queue, ScheduledThreadPoolExecutor, and CompletableFuture to batch database queries, reduce connection overhead, handle high concurrency, and includes full Java code examples and performance testing.
This report analyzes performance issues in a logistics platform, identifying backend IC card interface timeouts causing database connection pool exhaustion and frontend problems from excessive DOM elements and network loading, then proposes optimizations including interface migration, CDN, compression, preloading, domain sharding, and virtual scrolling.
This guide explains how to edit the nginx.conf file to configure the number of worker processes, determine appropriate values based on CPU cores and workload, and bind workers to specific CPU cores using worker_cpu_affinity, including practical command examples and considerations for blocking I/O.
This article outlines the essential design principles, client‑side optimizations, CDN usage, clustering, caching, database tuning, and service governance techniques required to build and maintain high‑traffic, high‑concurrency backend systems that are reliable, scalable, and performant.
Discover a comprehensive set of ten practical techniques—including batch processing, asynchronous execution, caching, pooling, parallelization, indexing, transaction management, and SQL tuning—to dramatically reduce API latency and improve efficiency in legacy backend systems, illustrated with real-world examples and code snippets.
This article outlines twelve practical techniques—including batch processing, asynchronous execution, caching, pre‑processing, pooling, parallelization, indexing, transaction sizing, code refactoring, pagination, SQL tuning, and lock granularity—to dramatically reduce API latency and improve overall backend efficiency.
This article outlines a comprehensive set of backend API performance optimization techniques—including batch processing, asynchronous execution, caching, pre‑processing, pooling, parallelization, indexing, transaction handling, SQL tuning, and lock granularity—to reduce latency and improve system efficiency.
This article presents a comprehensive guide to optimizing API latency by covering batch processing, asynchronous execution, caching, pre‑processing, pooling, parallelism, indexing, transaction management, code refactoring, pagination, SQL tuning, and lock granularity, each illustrated with practical code examples and diagrams.
This article explains why latency analysis is crucial for modern online services, compares common techniques such as RPC monitoring, CPU profiling, and distributed tracing, and then details the principle, implementation, and real‑world impact of critical‑path analysis in large‑scale distributed systems.
This article details the design and optimization of Jimu’s card‑based stranger‑social recommendation system, covering product background, gameplay flow, technical challenges in strategy and engineering, a multi‑task AI ranking model, vector recall improvements, and the resulting performance gains.
This article presents a comprehensive case study of how JD.com tackled the growing load of its shopping‑cart service by introducing an asynchronous pagination mechanism for auxiliary product information, detailing the background, objectives, technical architecture, challenges, and measurable performance and cost benefits.
This article examines a high‑traffic video‑watching scenario, identifies database write bottlenecks, and presents four asynchronous solutions—thread‑pool, in‑memory queue with scheduled tasks, message‑queue, and Agent‑plus‑MQ—detailing their implementation steps, trade‑offs, and best‑practice guidelines.
This article details a comprehensive performance‑testing and optimization effort for a high‑traffic site‑service system, covering hardware setup, CAT monitoring integration, bottleneck identification, Spring Gateway and Netty tuning, asynchronous logging, and the resulting dramatic improvements in QPS and response times.
This article explores why asynchronous processing is crucial for high‑concurrency scenarios and compares five practical solutions—including thread‑pool models, local memory with scheduled tasks, MQ patterns, and Agent‑plus‑MQ—to reduce write latency and improve system throughput.
This article explains the motivation behind using filters to reduce database I/O, introduces Bloom filters and their bitmap implementation, discusses their limitations, then covers Counting Bloom filters and Cuckoo filters—including their hash structures, deletion support, performance trade‑offs, and practical improvements.
This article presents a comprehensive performance testing case study of a multi‑table database query interface, detailing common performance metrics, response‑time classifications, step‑by‑step load‑test results, root‑cause analysis of CPU, memory, and connection‑pool bottlenecks, and demonstrates how tuning Druid’s maxActive and maxWait parameters optimizes throughput and reduces latency.
Alibaba’s display advertising engine ‘Transformers’ introduces a four‑generation compute‑efficiency framework that allocates compute dynamically based on request value and controls real‑time latency through a three‑step, feedback‑driven process, delivering deeper retrieval, lower P99 latency, higher CTR and RPM while maintaining stable recall‑chain performance.
This article explains how caching can accelerate read/write performance and reduce backend load, analyzes its benefits and costs, and details practical design patterns such as update policies, granularity control, penetration, bottom‑hole, avalanche, and hot‑key optimizations for reliable high‑performance systems.
This article presents a comprehensive study of local vector retrieval optimization, covering memory‑based filtering techniques, Redis‑backed vector storage designs, and various sorting algorithms—including radix and heap‑based approaches—to achieve lower latency and higher throughput for large‑scale ad recommendation systems.
The article explains why caching is essential in high‑concurrency distributed systems, analyzes its benefits and costs, and then details various cache update, penetration, no‑hole, avalanche, and hot‑key rebuilding techniques, offering practical guidance for backend engineers.
Baidu’s Short Video Push System is a distributed platform serving hundreds of millions of users across multiple apps, delivering personalized, real‑time notifications via a modular architecture that includes material and user centers, recall, preprocessing, and delivery services, while optimizations such as activity‑based scheduling, bitmap‑based user segmentation, consistent‑hash frequency control, and protobuf compression boost click‑through rates, scalability, and resource efficiency.
This article provides a step‑by‑step guide on optimizing SpringBoot projects by modifying configuration files and tuning JVM parameters, covering both IDE‑based VM options and command‑line scripts, with detailed explanations of each setting and practical examples.
This article presents a comprehensive case study of the technical challenges and optimization strategies—including traffic surges, cache penetration, cache breakdown, limit‑purchase handling, and inventory deduction—encountered during a large‑scale ticket reservation event, and demonstrates how systematic backend improvements achieved over 50% performance gains and 98% cache hit rate.
This article explains how high‑throughput e‑commerce order processing can suffer from inventory oversell when using a single distributed lock, analyzes the lock’s concurrency bottleneck, and proposes a segment‑locking optimization—splitting stock into multiple lock keys—to dramatically increase parallelism while addressing edge‑case pitfalls.
Volcano Engine RTC shares its best‑practice solutions for large‑scale interactive scenarios—including thousand‑person audio chat, live‑stream co‑hosting, and cloud rendering—detailing server‑side stream selection, intelligent mixing, and edge‑node integration to reduce latency, bandwidth, and computational load while enhancing user experience.
The iQIYI International backend team boosted global video delivery by introducing a web‑cache layer, Brotli compression, and a two‑level Caffeine‑Redis caching system with pub/sub synchronization, achieving up to 6% faster First Contentful Paint, near‑100% cache‑hit rates, and measurable revenue gains.
This article walks through the complete troubleshooting and performance‑tuning process for a high‑concurrency SSM‑based e‑commerce system, covering problem identification, root‑cause analysis, JVM, Tomcat, Redis and MySQL optimizations, traffic shaping, and the resulting stability improvements.
Learn how to overcome the concurrency bottleneck of traditional distributed locks in high‑traffic e‑commerce scenarios by applying segment‑based locking and other optimization techniques, enabling thousands of orders per second while preventing inventory overselling.
The article examines the technical difficulties of handling tens of thousands of requests per second in flash‑sale systems, analyzes root causes such as API design, overload, cheating, and data‑safety issues, and proposes practical backend strategies—including Redis caching, load‑balancing, rate‑limiting, and optimistic locking—to maintain performance and reliability.
The article explains how to reduce redundant service calls in Java back‑end systems by passing required data from upper layers, applying cache annotations, and implementing an elegant ThreadLocal‑based caching mechanism using custom annotations, aspects, and filters to improve performance without extensive code changes.
This article details the background, performance and stability requirements, strategic approach, and concrete initiatives—including full‑chain load testing, chaos engineering, monitoring, and targeted optimization projects—that were undertaken to boost the performance by over 300% and improve high‑availability of the continuation enrollment platform.
This article details a step‑by‑step performance tuning process for an SSM‑based e‑commerce system experiencing severe CPU, memory, Redis, and JDBC issues during flash‑sale spikes, covering root‑cause analysis, JVM and Tomcat optimizations, connection‑pool adjustments, code refactoring, and the resulting stability improvements.
Xianyu improved its offline‑push delivery by fixing duplicate Agoo calls, using device IDs, classifying messages to avoid vendor quotas, prompting notification permissions, and personalizing timing and content, which raised the weak‑perception delivery rate by double‑digit percentages and set the stage for path tracing and infrastructure decoupling.
This article details how Ctrip's hotel search system was optimized through storage compression, spatial indexing, KV storage, semantic query generation, context‑aware error correction, and a custom domain‑specific language, balancing performance, flexibility, and user experience for large‑scale online travel services.
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.
The article outlines five key layers—front‑end optimization, middle‑layer load distribution, gateway control, service‑layer improvements, and database tuning—detailing practical techniques such as static resource caching, Nginx load‑balancing methods, request rate limiting, message‑queue buffering, and extensive MySQL best‑practice guidelines to build robust high‑concurrency architectures.
After receiving a Young Generation GC alarm on a Pinduoduo service, I traced the issue to a high‑frequency scheduled task that created massive objects, then resolved it by breaking the job into finer‑grained tasks and finally sharding the work across multiple machines.
This article analyzes the challenges of scaling group chat in an instant messaging system and presents architectural optimizations—including shared message storage, periodic conversation list updates, offline count handling, and version‑based incremental sync—to reduce write and read amplification while improving overall performance.
This article examines the technical challenges of handling massive concurrent traffic during flash‑sale events, explains why backend APIs become bottlenecks, and presents practical strategies—including request design, overload protection, anti‑abuse measures, and data‑safety techniques such as pessimistic, FIFO, and optimistic locking—to keep web systems stable and performant.
This article describes how Ctrip's vacation price engine evolved from a single‑threaded 1.0 version to a highly parallel 3.0 version, detailing the business scope, core challenges, architectural redesigns, resource‑specific optimizations, and the resulting improvements in accuracy and computation speed.
This article details a competition‑level solution that implements a high‑throughput HTTP server using a custom multi‑threaded epoll model, UDP‑based protocol conversion, sparse indexing, and optimized write‑read schemes to achieve top performance on limited‑resource hardware.
This article explores the technical challenges of handling massive concurrent requests in e‑commerce flash‑sale and ticket‑booking systems, covering backend interface design, performance bottlenecks, overload protection, cheating mitigation, and data‑safety techniques such as optimistic locking and queueing.
This article outlines a comprehensive, step‑by‑step approach to performance testing and root‑cause analysis for backend services, covering environment validation, tool selection, Linux system limits, dependency checks, empty‑endpoint verification, throughput calculation, log monitoring, and essential Linux commands such as netstat, vmstat, mpstat, iostat, top and free.
This article examines the technical challenges of handling 50,000 requests per second in flash‑sale systems, covering request interface design, high‑concurrency bottlenecks, overload protection, anti‑cheat mechanisms, and data‑safety techniques such as Redis caching, pessimistic and optimistic locking.
This article examines the technical challenges of handling massive concurrent requests during flash‑sale and ticket‑buying events, offering practical backend design, performance tuning, anti‑cheat measures, and data‑safety techniques to keep web systems stable under extreme load.
This article explains the fundamental concepts behind Nginx web caching, covering the proxy‑store and resource‑not‑found mechanisms, configuration examples, and how memcached can be integrated to efficiently store and retrieve cached responses, improving performance and resilience.
This article describes how adjusting the PHP‑FPM max_requests setting and modifying its source code can smooth CPU idle and memory usage fluctuations in a high‑traffic online service, improving resource stability and reliability.
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 explains how Taobao built a high‑traffic flash‑sale platform by isolating hot data, separating static and dynamic content, applying time‑slice throttling, real‑time hotspot detection, and numerous backend optimizations such as Java concurrency tuning, local caching, queueing and database‑level row‑lock scheduling.
This article examines the technical difficulties of e‑commerce flash‑sale and ticket‑buying spikes, explains why massive concurrent requests strain web back‑ends, and presents practical design, overload‑protection, anti‑cheat, and data‑safety strategies such as CDN, Redis, locking, and IP filtering.
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.
The article explains why modern web systems face ever‑increasing concurrent connections, analyzes how front‑end techniques and server‑side configurations such as Apache MPM modes, Nginx, sendfile, and epoll can reduce memory and CPU consumption, and offers practical recommendations for efficient backend architecture.
This article explains why DNS resolution slows down mobile app traffic, compares the traditional PC‑era HTTP flow with wireless app challenges, and presents a DNS‑free optimization that uses a locally cached IP list to improve latency, load balancing, and resilience.
This article examines why modern web systems face exploding concurrent connections, explains how richer pages and browser limits increase load, and presents front‑end caching, request merging, Apache/Nginx memory and CPU optimizations, and practical guidelines for reducing server resource consumption.
This article explores the challenges of handling tens of thousands of requests per second in flash‑sale systems, covering interface design, QPS calculations, overload protection, anti‑cheat measures, and data‑safety techniques such as pessimistic, optimistic, and queue‑based locking.
