This article presents a practical Python benchmark that measures token‑per‑second generation speed of various large language models—including GPT‑4o, glm‑4‑airx, and moonshot‑v1‑32k—by timing text generation on a Colab environment and summarizing the results in detailed tables and visual charts.
Mako is a lightning‑fast, zero‑configuration web bundler whose Rust core and parallel Node.js pool handle loading, parsing, transforming, dependency analysis, and asset generation across JavaScript/TypeScript, CSS, assets, and more, while offering HMR, code‑splitting, plugin extensibility, and production‑grade reliability.
This article introduces Python coroutines, compares them with multithreading, explains when to use them, and demonstrates how the httpx library can perform asynchronous HTTP requests to dramatically improve backend performance, including installation steps and a benchmark against synchronous requests.
This article introduces a collection of practical open‑source utilities—including SQLTXPLAIN, SOAR, SQLAdvisor, and Percona Toolkit tools—that streamline common MySQL DBA tasks such as schema changes, replication checks, slow‑query analysis, and data consistency verification, helping engineers work faster and more efficiently.
This article explains Linux network‑stack enhancements—RSS, RPS, RFS, accelerated RFS, and XPS—detailing their purpose, configuration steps, and recommended settings to improve parallelism and latency on multi‑CPU systems.
This article explains how to identify and resolve the severe startup latency of a SpringBoot service by analyzing the SpringApplicationRunListener and BeanPostProcessor stages, narrowing scan packages, manually registering beans with JavaConfig, and handling cache auto‑configuration to reduce launch time from minutes to seconds.
This article explains how the LIMIT clause can cause massive I/O and CPU overhead in MySQL queries, analyzes the underlying B+‑tree index structure and execution plans, and presents practical optimization techniques such as covering indexes, sub‑queries, and partitioning.
This article explains how Baseline Profiles improve Android app start‑up performance, analyzes three major obstacles—unsupported app stores, dex‑crc mismatches caused by code‑obfuscation, and hot‑fix dex incompatibility—and provides concrete solutions including active optimization, dex‑name correction tasks, and profile injection during patch synthesis.
Nvidia plans to shorten its GPU generation cycle to one year, launching Blackwell Ultra in 2025, Rubin in 2026, and Rubin Ultra in 2027, while boosting token‑generation efficiency and introducing AI‑optimized Ethernet solutions like Spectrum‑X800, aiming to dominate large‑scale AI clusters and reshape the high‑performance networking market.
This article explains Netty's memory pool architecture after switching from jemalloc3 to jemalloc4, detailing the new memory size classes, core components such as PoolArena, PoolChunkList, PoolChunk, and PoolSubpage, and provides Java code snippets to illustrate their implementations.
The article explains how encapsulating peripheral registers in a struct rather than using separate global variables reduces literal‑pool usage, cuts instruction count, and improves execution speed on Cortex‑A9, providing step‑by‑step assembly analysis, compilation commands, and further optimization techniques.
This article compares Memcached, Redis, and Aerospike across persistence, replication, performance, memory management, and use‑case scenarios to help you select the most suitable in‑memory database for high‑throughput, low‑latency applications.
This article explains how Redis achieves tens of thousands of QPS on average hardware, breaks down the technical reasons behind its speed—including C implementation, in‑memory design, epoll‑based I/O multiplexing, and a single‑threaded event loop—while also detailing the evolution to multithreaded I/O in later versions.
This article investigates MySQL's recommendation against using UUID or non‑sequential random IDs as primary keys by benchmarking auto‑increment, UUID, and Snowflake‑derived keys, analyzing their insertion speed, index behavior, and discussing the advantages and drawbacks of each approach.
This article explains the Nginx location matching hierarchy—including exact, longest‑prefix, regex, ^~, and default matches—illustrates each rule with configuration examples, and offers practical recommendations for writing efficient and reliable server blocks.
This article details the architecture, business value, configuration, storage choices, performance challenges, and future plans of Bilibili's live streaming ranking system, explaining how it uses MySQL and Redis, supports automated scoring, and handles high‑traffic read/write demands.
This article explains Android ANR (Application Not Responding) fundamentals, detailing its four types, underlying causes such as main‑thread blocking and deadlocks, the system’s timeout mechanisms, step‑by‑step analysis workflow, trace interpretation, and practical best‑practice tips to prevent and resolve ANR problems.
This article walks through Java 17’s most useful language upgrades—including text blocks, enhanced NullPointerException messages, records, switch expressions, private interface methods, pattern matching, collection factories, Stream API extensions, the new HttpClient, JShell, direct file execution, and ZGC—showing why upgrading from Java 8 can boost code clarity and productivity.
This article uses a realistic interview scenario to explore why multi-table joins in MySQL are not inherently slow, detailing four join algorithms—simple nested-loop, index nested-loop, block nested-loop, and batched key access—along with their pseudo‑code, execution steps, performance trade‑offs, and the impact of features like MRR and recent optimizer changes.
This article analyzes a MySQL query that originally took nine seconds, identifies redundant joins and an EXISTS subquery on the same table, and demonstrates how restructuring the query eliminates duplicate scans, reducing execution time to 0.2 seconds.
This article explains the motivations for creating a custom PHP framework—such as meeting specific project needs, deepening technical understanding, ensuring long‑term maintainability, optimizing performance, improving usability, and enhancing security—while outlining the curriculum of the accompanying tutorial series.
This article explains the concept of false sharing caused by CPU cache line interactions, demonstrates how cache padding can mitigate the performance penalty, and provides Go benchmark code and results to illustrate the impact on multi‑core concurrency.
This article walks through practical Java backend performance techniques—including parallel processing with CompletableFuture, fine‑tuned thread‑pool configuration, transaction scope minimization, cache‑line awareness, object‑pool usage, lock granularity, copy‑on‑write collections, and network payload reduction—backed by concrete code samples, benchmark results, and step‑by‑step analysis of trade‑offs and best practices.
The 2023 NetEase Cloud Music annual‑report front‑end case study details how sub‑second first‑screen loads, SPA routing with TypeScript, GPU‑accelerated animations, optimized media handling, multi‑layer quality monitoring, and a unified development platform together boost performance, reliability, and engineering efficiency, driving higher DAU and share‑rate.
This article explains why developers—both frontend and backend—should understand browsers, covering user experience, compatibility, performance, security, and debugging, then traces the browser’s evolution from the early 1990s to modern multi‑process architectures, outlines core components, rendering engines, and current market share.
A comprehensive benchmark compares multiple PHP runtimes—including Apache, Nginx, Nginx Unit, Roadrunner, FrankenPHP, and Swoole—under identical Docker‑based test conditions using k6 load testing, revealing significant performance differences and offering practical recommendations for Symfony applications.
The article analyzes why deploying MySQL in Docker containers can lead to data‑security, performance, state‑management and resource‑isolation problems, while also outlining scenarios where containerizing MySQL might be feasible and offering practical mitigation strategies.
This article walks through a real‑world Redis latency incident, detailing systematic troubleshooting steps, key metrics to monitor, command‑line diagnostics, and practical optimizations such as scaling, pipelining, and hot‑key mitigation to restore service performance.
This article explains why Redis achieves extremely high performance, reaching up to 100,000 QPS, by leveraging its in‑memory design, I/O multiplexing, optimized data structures such as SDS, ziplist and skiplist, and a single‑threaded event loop, each detailed with examples and code.
When multiple CentOS 7 servers in the same rack experience 30‑60 second delays before prompting for an SSH password, the slowdown is usually caused by reverse DNS lookups and GSSAPI authentication, which can be resolved by disabling UseDNS and GSSAPIAuthentication in sshd_config and restarting the service.
The article explains how React 19 silently altered Suspense behavior by disabling parallel rendering of sibling components, turning data fetching into a waterfall pattern, which caused significant performance regressions for many sites, sparked community backlash, and was eventually rolled back after extensive discussion.
This article explains Java heap memory concepts, garbage collection mechanisms, common memory issues, and outlines the team's quality assurance framework, including testing left‑shift with AI code review and right‑shift monitoring to improve reliability and performance.
This article details Didi’s use of Elasticsearch 7.6, covering its architecture on physical machines, gateway and control layers, user console features, deployment models, data synchronization strategies, engine iterations such as fine‑grained tiered protection, multi‑active replication, JDK 17/ZGC performance upgrades, cost reductions, multi‑tenant isolation, security enhancements, and future migration to ES 8.13.
PHP’s rise to the 15th spot in the TIOBE June ranking highlights its enduring strength in web development, driven by mature frameworks, performance gains, a vibrant community, and integration with modern tools like Docker and front‑end frameworks, while also facing future challenges in scalability and cloud‑native adoption.
This article explains how Ceph cache pools work, compares read‑only and write‑back cache types, provides step‑by‑step commands for creating and configuring cache pools, adjusting CRUSH rules to target SSDs, and outlines safe removal procedures for both cache modes.
This article introduces JMH, explains why it outperforms simple loops or other tools, and provides step‑by‑step Maven setup, benchmark creation, execution, and advanced annotations such as @Warmup, @Fork, @Setup, Blackhole usage, and SpringBoot integration for accurate Java micro‑benchmarking.
This article explains Java virtual threads introduced in Java 19, compares them with traditional OS threads, shows how to enable them in Spring Boot, and presents performance benchmarks that demonstrate dramatic latency reductions for I/O‑bound workloads.
This article investigates whether defining a VARCHAR column with a larger length affects MySQL storage space and query performance by creating two tables with VARCHAR(50) and VARCHAR(500), inserting one million rows, and measuring storage usage and execution times for various queries.
This article presents ten practical Java 17 tips—ranging from using Optional and String.valueOf() to leveraging Arrays.copyOf, isEmpty(), pre‑compiled regex, and generics—to avoid common pitfalls, improve performance, and write more readable and robust backend code.
This article explores how the depth of a webpage's DOM tree influences memory usage, style calculations, and rendering performance, presenting experiments that compare shallow and deeply nested structures and offering practical tips for monitoring and optimizing DOM depth.
The article discusses a top architect's interview with a game company about microservice adoption, explains why real‑time performance and stateful requirements often make microservices unsuitable for game servers, and presents multiple expert opinions highlighting latency, network overhead, and team size considerations.
This article explains how Java's parallelStream uses the shared ForkJoinPool, why developers may need to customize the thread pool for better performance and isolation, and provides two approaches—setting the system property and creating a dedicated ForkJoinPool—along with complete code examples and execution results.
This article explores essential Linux I/O techniques, covering basic file operations with system calls like open, read, write, and close, advanced methods such as memory‑mapped files and asynchronous I/O, network programming with sockets and multiplexing, plus performance‑tuning strategies for efficient resource utilization.
To handle high‑traffic web workloads, the default Linux kernel settings must be adjusted; this guide explains the key sysctl parameters—such as file‑max, tcp_tw_reuse, tcp_keepalive_time, and buffer sizes—and how to apply them for optimal Nginx concurrency.
This article explains the common slow approach of reading Excel files with POI and inserting rows via JDBC, then presents three practical optimizations—caching data, using asynchronous multithreaded reads, and per‑file async processing—along with thread‑pool configuration tips and EasyExcel examples to dramatically reduce import time from minutes to seconds.
This article details Zhihu's migration of massive Spark and MapReduce shuffle workloads from the External Shuffle Service (ESS) to a push‑based Remote Shuffle Service (RSS) powered by Celeborn, covering background problems, evaluation of open‑source implementations, deployment architecture, encountered issues, solutions, performance gains, and future plans.
Claude 3.5 Sonnet, Anthropic’s latest large language model, is evaluated across a range of Chinese‑language tasks, visual reasoning, coding, and game creation, showing faster, cheaper, and often superior results compared to GPT‑4o, while also revealing occasional failures in simple games and math problems.
This article explains the subtle differences between the hostname localhost and the IP address 127.0.0.1, covering their definitions, resolution processes, performance impact, IPv6 considerations, practical usage scenarios, and best‑practice recommendations for developers.
This article details the motivation, design, implementation, testing, and rollout of a refactored Kubernetes DNS solution that replaces the default kube-dns → CoreDNS chain with a node‑local q‑dnsmasq cache and parallel upstream queries to achieve higher availability, faster resolution, and better cache hit rates in large‑scale clusters.
This article provides a detailed comparison of InfiniBand and RoCEv2 network architectures, examining their technical features, flow‑control mechanisms, performance, cost, and suitability for AI compute environments to guide designers in selecting the optimal solution.
The article measures the exact time cost of establishing and closing a MySQL connection from Java, breaks down each network and protocol step with Wireshark captures, and shows how these latencies add up to hundreds of milliseconds, making connection pooling essential for high‑traffic web services.
This article explains the cache‑stampede problem when hot Redis keys expire under massive traffic, illustrates why simple caching fails, and presents a mutex‑lock solution with Go code to safely rebuild caches while protecting the database.
To keep a WeChat mini‑program under the 2 MB limit and improve both startup and runtime performance, the article proposes splitting the main bundle into async sub‑packages, removing unused code, pre‑fetching data, using on‑demand imports, and implementing a backend‑driven size‑control and CI release gate.
Learn how to boost Node.js application performance using the popular lru-cache package, covering the LRU algorithm fundamentals, basic installation and usage, advanced features like peek, fetchMethod, and custom disposal, plus an in‑depth look at its internal Map and doubly‑linked list implementation.
This article explains the concept of API gateways, compares popular solutions such as Kong, Zuul, Spring Cloud Gateway and Tyk, details Kong's architecture, features, performance benchmarks, deployment scenarios, monitoring setup, and the practical problems encountered with their resolutions.
The MoonWebTeam Frontend Tech Monthly curates the latest front‑end frameworks, performance optimization case studies, AI innovations, domain‑driven design insights, and practical best‑practice techniques, offering developers a comprehensive snapshot of cutting‑edge web development trends.
This article investigates the mysterious "prevent gc" comment in RocketMQ source code, explains how Thread.sleep(0) interacts with JVM safepoints and counted loops, and demonstrates practical fixes by changing loop indices or inserting explicit safepoint calls.
At PyCon 2024, experts demonstrated how static typing, Cython compilation, static‑linked C extensions, and sub‑interpreter architectures can dramatically accelerate Python code, reduce runtime overhead, and mitigate GIL limitations, offering practical pathways for faster, more scalable Python applications.
This article walks through a series of practical performance‑boosting techniques for Vue‑based web applications, covering Webpack bundle analysis, Gzip compression, WebP image conversion, HTTP/2 configuration, Nginx tweaks, and OSS/CDN integration, with concrete code snippets and step‑by‑step guidance.
This article compares Google’s Protocol Buffers and the Swagger/OpenAPI framework, detailing their features, data formats, use cases, flexibility, development workflows, and why Google created protobuf, helping developers choose the right tool for high‑performance data serialization or comprehensive API documentation.
The article analyzes Nvidia’s historic advantage, the massive performance and energy efficiency gains from Pascal to Blackwell GPUs, the economics of training large language models like GPT‑4, and the detailed roadmap of upcoming GPU, memory, and interconnect technologies shaping the future of data‑center AI.
This article explains common scenarios that cause MySQL or Java processes to consume 900% CPU, walks through step‑by‑step diagnosis using Linux tools, and provides concrete optimization techniques such as indexing, caching, thread analysis, and code adjustments to bring CPU usage back to normal levels.
This article examines the performance differences between using the '+' operator and StringBuilder for string concatenation in Java, presenting JUnit benchmark tests for simple concatenations and looped concatenations, analyzing compiled bytecode, and concluding that '+' is suitable for simple cases while StringBuilder excels in loops.
This article compiles seven hands‑on Linux performance optimization experiences, covering key performance factors, rapid troubleshooting steps for CPU, memory, disk and network issues, load‑average analysis, top‑process identification, memory‑stat discrepancies, swap usage scenarios, and detailed TCP kernel tuning recommendations.
The author benchmarked Go's standard HTTP library against PHP frameworks Workerman, Webman, and Swoole (including coroutine mode) on a 4‑core Ubuntu 20.04 Aliyun instance using ApacheBench with and without keep‑alive, revealing that Workerman and Webman often outperform Go in raw request‑per‑second throughput.
This guide introduces Memray, a powerful Python memory analysis tool, explains how to install it, run command‑line profiling, integrate it into code with the Tracker API, interpret detailed reports, and apply advanced filtering and merging techniques while noting its runtime overhead.
This article explains Hugging Face's KV cache quantization technique, detailing how it reduces memory usage for long‑context LLM generation, the underlying quantization methods, implementation steps in 🤗 Transformers, benchmark results versus fp16, and the trade‑offs between speed, memory, and accuracy.
This article presents twelve practical guidelines for Java logging, covering interface‑layer usage, avoiding separator lines, preventing logging‑induced exceptions, leveraging Fastjson features, preserving exception stacks, limiting message length, merging stack traces, skipping costly placeholders, avoiding console output, discarding redundant utilities, and adhering to Alibaba's logging specification.
Traditional LAMP stacks, relying on php‑fpm and nginx, suffer from synchronous blocking and high CPU usage, limiting throughput for IO‑intensive microservice workloads, while benchmarks show that upgrading to PHP 7 and using a modern, asynchronous PHP framework like Webman can dramatically improve QPS and reduce latency.
The article examines Peter Zaitsev’s critique of Oracle’s stewardship of MySQL, debunks long‑standing myths, evaluates recent MySQL Heatwave features and performance gaps, and compares MySQL’s trajectory with PostgreSQL and other open‑source databases.
This guide compiles practical Nginx performance‑tuning directives, FastCGI settings, Linux kernel parameter adjustments, security hardening tips, and monitoring configurations, providing concrete examples and command‑line snippets to help sysadmins and developers optimize high‑concurrency web services.
This article analyzes MySQL's recommendation to avoid UUID or non‑sequential random primary keys by benchmarking insert performance across three tables—auto_increment, UUID, and random key—explaining the underlying index structures, observed speed differences, and the trade‑offs of each approach.
This article examines four common WeChat red packet distribution algorithms—simple random, double‑mean, line‑segment division, and array_rand optimization—explaining their principles, PHP implementations, performance characteristics, and how to verify randomness, providing code samples and comparative analysis for developers.
This article explains how to design a MySQL table for product sales statistics, initialize massive test data, and compare several query strategies—including loop queries, OR‑concatenated dynamic SQL, mixed filtering, and the SQL‑92 row‑comparison technique—to efficiently retrieve sales figures for multiple business units and their products without violating development constraints.
This guide introduces Uber’s open‑source Zap library for Go, explains its performance‑focused features, shows how to install and configure it, and provides step‑by‑step code examples for basic logging, sugared logging, level control, file output, and log rotation.
This article analyzes the performance bottlenecks of an H5 map built with the AMap JS‑SDK, explains the causes of frame drops on low‑end devices, and presents a step‑by‑step refactor using data‑fetching tweaks, DistrictCluster aggregation, and conditional rendering to achieve smoother interactions.
Microsoft’s Windows now runs on Qualcomm’s ARM processors through a combination of native kernel adaptations, driver and library recompilation, and an x86‑to‑ARM emulation layer, balancing performance gains from direct ARM execution with compatibility‑driven overhead, while outlining ongoing optimization efforts and future prospects.
The article explains how Fast‑Retry, an asynchronous multi‑task retry framework, dramatically outperforms synchronous retry libraries like Spring‑Retry and Guava‑Retry when handling massive workloads, and provides step‑by‑step code examples, performance benchmarks, and integration tips for Java developers.
This whitepaper examines the industry trend of moving data‑intensive analytics workloads to cloud‑native platforms, analyzes the unique cost model of cloud storage, and presents case‑study findings from Uber's Presto production environment to guide efficient I/O design and cost‑effective performance optimization.
The article explores how Nordic programmers like Linus Torvalds, Andreas Kling, and Ville Turjanmaa have built full operating systems—from Linux to SerenityOS and the assembly‑only MenuetOS—highlighting their motivations, technical achievements, performance gains, and the cultural environment that fuels such ambitious open‑source projects.
Tencent News upgraded its recommendation architecture by consolidating data platforms, redesigning index and feature services, adopting DDD and Lambda/Kappa patterns, and adding robust debugging and stability measures, which boosted availability to 99.99%, cut CPU, memory and cost by over 60%, and accelerated development and experiment cycles.
This article explains Java virtual threads, compares them with traditional OS threads, shows how to enable them in SpringBoot 3, and presents performance benchmarks demonstrating up to 200‑fold speed improvements for I/O‑bound workloads.
This article compares Java's standard ThreadLocal with Netty's FastThreadLocal, explains their internal mechanisms, shows practical code examples, runs a JMH microbenchmark, and discusses performance, memory management, and usage scenarios to help developers choose the right thread‑local storage solution.
This article introduces several MySQL performance‑diagnostic utilities—including mysqltuner.pl, tuning‑primer.sh, pt‑variable‑advisor, and pt‑query‑digest—explains how to download and run them, interprets their reports, and also contains promotional material for unrelated AI services.
The article walks through real interview scenarios where MySQL or Java processes consume 900% CPU, detailing step‑by‑step diagnosis with top, processlist, jstack, and concrete remediation such as adding indexes, using caches, fixing busy loops, and tuning memory.
The article defines large database transactions, explains their performance and replication risks, offers practical guidelines to avoid them, demonstrates a MySQL example that creates and repeatedly inserts data to illustrate growing execution time, and discusses challenges and solutions for handling massive tables in high‑traffic environments.
An in‑depth postmortem of a costly MySQL data‑migration failure caused by using INSERT INTO SELECT without proper indexing, detailing the OOM issue, full‑table scans, lock contention, and best practices to safely employ the statement.
FrankenPHP, a Go‑based PHP application server built on Caddy, claims to double web request performance and halve developer/DevOps workload, offering container‑ready deployment, automatic HTTPS, native PHP extensions, in‑memory mode, structured logging, Prometheus metrics, and experimental HTTP/3 support.
This article outlines seven typical cache problems—including cache penetration, breakdown, avalanche, large keys, hot keys, hit‑rate issues, and data inconsistency—and provides practical solutions such as parameter validation, Bloom filters, null‑value caching, locking, auto‑renewal, random expiration, high‑availability setups, compression, and cache warming to improve system reliability and performance.
This article explains the concept of the Virtual DOM, its performance benefits in front‑end development, and presents several open‑source implementations such as Million.js, Snabbdom, virtual‑dom, blockdom, Maquette, and additional lightweight libraries, each with brief descriptions and GitHub links.
This article dissects Pulsar’s storage layer by examining the Bookie component, detailing its architecture, the sequential write‑ahead log, ledger management, journal handling, caching mechanisms, and the complete read/write call chains with concrete code examples and performance considerations.
This article investigates two real‑world performance incidents—one caused by excessive disk I/O from a misconfigured Filebeat and another by kernel memory‑fragmentation bugs triggered by a trace feature—detailing observations, Linux diagnostic commands, analysis, and practical remediation steps.
GraalVM Native Image now supports Native Memory Tracking (NMT), allowing developers to monitor off‑heap memory allocations, compare Java and native execution, integrate with JFR, and assess performance impact through detailed reports and experimental events.
This article explains how to use a custom CommonContext with conditional logging and lambda‑based delayed evaluation to quickly trace issues in high‑concurrency Java back‑end systems without degrading performance, and demonstrates the performance improvements through before‑and‑after measurements.
This article documents the author's practical migration from JDK 8 to JDK 11, covering motivations such as performance, compatibility, and industry trends, detailed upgrade steps, GC tuning, resolved issues, and showcases significant performance gains and new Java 11 language features with code examples.
This article explains database sharding concepts, reasons for using it, various sharding strategies, and demonstrates implementation with ShardingSphere, including configuration and Java code examples, while also offering additional resources for advanced architecture and interview preparation.
This article compares the Go web frameworks Echo and Gin, detailing their main features, performance, API design, middleware and routing capabilities, community support, and provides side‑by‑side code examples to help developers choose the most suitable framework for their projects.
Zero‑copy is an optimization technique that eliminates unnecessary memory copies between kernel and user space by leveraging DMA, memory‑mapping and specialized system calls, thereby reducing CPU load, latency and improving throughput for high‑performance networking, storage and multimedia workloads.
This article introduces Python decorators tailored for mathematical functions, presenting four practical examples—a timing decorator, a simple automatic differentiation decorator, a result‑caching decorator, and a random‑noise decorator—each accompanied by complete code snippets and explanations.
This article explains the essential conditions, metrics, and scenario classifications for building high‑concurrency systems, then details common solutions such as database read/write separation, local and distributed caching, cache‑eviction policies, handling master‑slave lag, preventing cache penetration and avalanche, and applying CQRS to achieve scalable, high‑performance back‑end services.
