By applying Dubbo's GenericService $invoke method, the vivo project replaced tightly‑coupled direct API calls with a unified remote‑call layer that eliminates third‑party JAR dependencies, reduces system coupling, and leverages Dubbo’s internal proxy‑exporter mechanisms to dynamically invoke diverse external services.
This article details the design and implementation of Baidu's transaction middle‑platform asset system, covering its C2C payment model, core asset management capabilities, consistency guarantees, double‑entry accounting, hot‑account handling, sharding and archiving strategies, and compliance with national financial regulations.
This article analyzes the challenges of real‑time inventory synchronization for omnichannel retail—such as sync latency, concurrent overselling, and shared stock calculation—and presents a comprehensive backend architecture featuring low‑latency syncing, dynamic throttling, safety‑stock algorithms, and a scalable calculation engine.
This article provides a comprehensive overview of microservice architecture, covering its definition, history, differences from monolithic development, key characteristics, comparison with SOA, practical implementation steps, common design patterns, advantages, disadvantages, and essential considerations for successful adoption.
This article explores the rapid growth of video traffic, explains why transcoding is essential, compares CPU, GPU, and VPU hardware for video processing, details the FFmpeg software stack, describes the design of a cloud‑native transcoding cluster, its scheduling, shard‑transcoding technique, and presents performance test results.
This article provides a concise overview of etcd and Kubernetes, detailing etcd’s features and use‑cases, explaining Kubernetes fundamentals, its relationship with Docker, and describing key components such as Minikube, Kubectl, Kubelet, common deployment methods, and the platform’s cluster management architecture.
This article examines common business scenarios that require timed execution, explains why scheduled tasks are essential, compares standalone and distributed Java scheduling frameworks—including Timer, ScheduledExecutorService, Spring, Quartz, TBSchedule, Elastic-Job, Saturn, and XXL-Job—and provides guidance on selecting the most suitable solution based on scalability, reliability, and operational needs.
This article analyzes Feakin's approach to online collaborative diagram editing, covering communication protocols, data formats, the trade‑offs between centralized OT and decentralized CRDT algorithms, and the Rust‑based server and client implementation using Actix, WebSocket, and WebAssembly.
The article outlines how a modest cluster of 20 virtual machines with 16 GB RAM each can handle 1 billion user records and 20 million hourly queries by using simple in‑memory arrays, basic indexing, modest bandwidth, and straightforward failover mechanisms, without requiring advanced technologies.
This article uses a simulated Taobao example to illustrate how a simple MVP monolithic architecture evolves through ten stages—from separating Tomcat and database to adopting caching, load balancing, database sharding, microservices, ESB, containerization, and cloud platforms—highlighting the technologies and design principles involved in each transition.
The article explains why e‑commerce users sometimes see their payments completed in the wallet but the order remains unpaid, analyzes internal and external drop‑order scenarios, and provides practical server‑side, client‑side, and active‑query techniques—including scheduled tasks and delayed‑message queues—to prevent such issues.
Rate limiting safeguards distributed systems by controlling request rates through algorithms such as leaky bucket, token bucket, fixed and sliding windows, and back pressure, while client‑side tactics like exponential backoff, jitter, and careful retries, and requires atomic distributed storage solutions (e.g., Redis+Lua) to avoid race conditions.
This article explains why order creation and payment APIs must be idempotent, outlines practical techniques such as unique request IDs, database unique constraints, Redis flags, and version‑based optimistic locking to avoid duplicate orders and solve the ABA problem in distributed environments.
ByteDance’s Cloud Shuffle Service (CSS) replaces the traditional Pull‑Based Sort Shuffle in Spark, FlinkBatch and MapReduce with a Push‑Based remote shuffle that improves stability, performance and elasticity, supports compute‑storage separation, and delivers significant speedups in large‑scale TPC‑DS benchmarks.
This article dissects why a well‑tuned Kafka cluster can process up to 20 million messages per second, examining producer batching and custom protocols, broker page‑cache, file layout and zero‑copy techniques, as well as consumer group strategies that together unlock its high throughput.
This article introduces essential distributed‑system concepts—including Bloom filters, consistent hashing, quorum, leader/follower roles, heartbeats, fencing, WAL, segment logs, high‑water marks, leases, gossip protocol, Phi failure detection, CAP and PACELC theorems, hinted handoff, read repair, and Merkle trees—explaining their purpose and how they are applied in systems such as BigTable, Cassandra, Dynamo, and Kafka.
This article explains how to guarantee that messages are never lost when using MQ middleware such as Kafka, RabbitMQ or RocketMQ, covering the three lifecycle stages, detection methods, design‑for‑failure principles, idempotent consumption, handling message backlog, and practical interview answers.
This article explains how to build a reliable delayed message queue using Redis, covering business scenarios, requirements such as persistence and retry, the design of Redis data structures, Lua scripts for atomic operations, and a Go implementation with example code for producing and consuming delayed tasks.
This article explains essential distributed‑system design patterns such as Bloom filters, consistent hashing, quorum, leader‑follower replication, heartbeats, fencing, write‑ahead logs, segment logs, high‑water marks, leases, CAP and PACELC theorems, hinted handoff, read‑repair, Merkle trees, and related failure‑detection mechanisms, illustrating how they improve scalability, consistency and fault tolerance.
This article explains the design and implementation of a unified data security control platform that protects user privacy and corporate data across multiple big‑data components (Hive, Hetu, GaussDB) by integrating Apache Ranger, custom authorization APIs, asynchronous processing, distributed locking, and SDK‑based authentication to achieve fine‑grained, one‑stop permission management.
This guide walks through MinIO client (mc) usage, bucket management, user and policy administration, and two practical methods for expanding a MinIO distributed cluster—peer‑to‑peer scaling and federation with etcd—providing step‑by‑step commands, scripts, and configuration details for cloud‑native object storage.
This article explains the Raft consensus algorithm in detail, covering its three node roles, leader election process, state‑machine log replication, handling of leader failures, multiple candidates, and split‑brain situations, providing clear diagrams and step‑by‑step descriptions for distributed system fundamentals.
This article explains how transactions, consistency, and consensus interrelate in distributed systems, reviews replication models, details ACID properties, internal consistency models, consensus algorithms, and Kafka’s configuration and transaction support, and highlights verification tools like TLA+ and Jepsen for ensuring correctness.
The article surveys three data‑replication models—master‑slave, multi‑master, and leaderless—explains how they enable scalability and fault‑tolerance, and examines core distributed‑system challenges such as partial node failures, unreliable networks, and unsynchronized clocks, while stressing safety‑liveness trade‑offs and design techniques like quorum and timeouts.
This article explains what search is, why traditional databases struggle with full‑text queries, introduces the concepts of inverted indexes and Lucene, and shows how ElasticSearch combines distributed architecture, real‑time analytics, and powerful search features to solve these problems.
This article traces RocketMQ's transformation from a traditional message queue to a cloud‑native, ultra‑fusion platform that integrates messaging, event‑driven architecture, and streaming, highlighting its historical stages, architectural upgrades, multi‑language SDKs, stateless consumption, and future cloud‑native trends.
This article explains the role of RocketMQ's NameServer, detailing its registration process, data structures, periodic heartbeat checks, failure cleanup, and how producers and consumers retrieve routing information to send and receive messages efficiently.
This article explains why delayed queues are needed for scenarios like unpaid orders or auto‑generated comments, compares built‑in and third‑party solutions, and provides a detailed design and implementation guide for a Redis‑based delayed‑queue service, including version‑1.0 features, version‑2.0 optimizations, and multi‑node deployment.
The article explains fundamental concepts, common use cases, and selection criteria for four popular message‑queue platforms—Kafka, RabbitMQ, Pulsar, and RocketMQ—detailing each system’s architecture, strengths, and drawbacks so readers can match workload, scalability, ordering, latency, and operational needs to the most suitable middleware.
This article provides an in‑depth overview of microservice concepts, RPC frameworks, serialization methods, distributed transaction models, CAP and BASE theories, monitoring solutions, high‑availability strategies, load‑balancing techniques, configuration management, service registration/discovery, and Dubbo fault‑tolerance clusters, offering practical examples and code snippets for Java‑based cloud native systems.
This article explores Go's concise design, strong concurrency model, fast compilation, static binaries, and broad ecosystem, highlighting its ideal fit for cloud‑native applications, distributed services, and command‑line tools while also addressing its limitations and future direction.
RedKV is Xiaohongshu's self‑developed NVMe‑SSD based distributed NoSQL KV store that combines Gossip and Shard control, delivering petabyte‑scale storage, near‑100 million QPS, multi‑cloud elasticity, low‑cost Redis compatibility, and advanced features such as token‑bucket rate limiting, online compression, backup‑read tail‑latency mitigation, and robust backup‑recovery mechanisms.
This article explains the evolution of software complexity, introduces microservice architecture as a modern solution, and details ten essential design patterns—including database per service, event sourcing, CQRS, Saga, BFF, API gateway, Strangler, circuit breaker, externalized configuration, and consumer‑driven contract testing—while outlining their advantages, disadvantages, and appropriate usage scenarios.
Reflecting on reading, the author reviews CAP theory’s consistency‑availability‑partition trade‑offs, extends ACID and BASE concepts, proposes modernizing CAP objects to consistency, fault and disaster tolerance, and examines how cloud‑native architectures, micro‑services, and SLA‑driven designs reshape fault tolerance and future self‑healing systems.
This article presents a concise overview of essential distributed‑system design patterns—including Bloom filters, consistent hashing, quorum, leader/follower, heartbeat, fencing, high‑water mark, leases, gossip protocol, PACELC theorem, hinted handoff, read‑repair, and Merkle trees—explaining their purpose, operation, and typical usage in large‑scale storage and coordination services.
This article introduces essential distributed‑system mechanisms—including Bloom filters, consistent hashing, quorum, leader/follower roles, heartbeats, fencing, write‑ahead logs, segment logs, high‑water marks, leases, gossip protocols, failure detection, CAP/PACELC theorems, hinted handoff, read‑repair, and Merkle trees—to help engineers design scalable and fault‑tolerant services.
This article details Ctrip's Data Replicate Center (DRC) architecture, challenges, and solutions for migrating MySQL databases to the cloud, covering read‑only and read‑write scenarios, cost analysis, proxy modules, transaction‑table replication, data isolation, filtering, deployment workflow, and achieved performance and cost benefits.
This article introduces graph database fundamentals, explains why graph databases are needed, outlines core storage goals such as index‑free adjacency, compares array, linked‑list and LSM‑tree storage schemes, and presents the design, performance advantages, and real‑world applications of the Galaxybase distributed graph database.
The article explains the concept of interface call idempotency in distributed micro‑service architectures, outlines scenarios that cause idempotency problems, and presents practical solutions such as token mechanisms, various lock strategies, unique constraints, deduplication tables, and global request IDs, with code examples in Java and Redis.
This article introduces the core concepts, components, exchange types, reliability mechanisms, ordering, delay, transaction, high‑availability, and load‑balancing strategies of three popular message‑queue systems—RabbitMQ, RocketMQ, and Kafka—while also discussing common challenges such as message ordering, delayed consumption, reliability, idempotence, and backlog handling.
This article explains the Chandy‑Lamport algorithm for capturing consistent global snapshots in distributed systems, describes its assumptions and message‑marker rules, walks through a detailed example with three processes and channels, and relates it to Apache Flink's asynchronous checkpoint mechanism.
This article explains the design, six mechanisms, and source‑code flow of Nacos’s Distro protocol, showing how asynchronous replication, periodic sync, new‑node loading, and local reads together provide AP‑style high availability and eventual consistency for service registration.
This article surveys common delayed‑message implementations—including external storage, RocksDB, Redis, and open‑source MQs like RocketMQ, Pulsar, and QMQ—detailing their architectures, advantages, disadvantages, and practical considerations for distributed asynchronous messaging systems.
This article provides a comprehensive overview of modern message‑queue systems, detailing RabbitMQ components and exchange types, RocketMQ’s core architecture and high‑availability features, and Kafka’s fundamental concepts, controller mechanisms, and consumer rebalance strategies, while also covering reliability, ordering, and dead‑letter handling techniques.
This article presents a comprehensive overview of NetEase's log collection and transmission platform, detailing its evolution from 2011 to the current Datastream‑NG architecture, the system's design goals, core component optimizations, operational monitoring, and future plans for intelligent scaling and diagnostics.
This article explains the fundamental components and mechanisms of RabbitMQ, RocketMQ, and Kafka—including exchanges, TTL, confirm and return listeners, consumer ACK/NACK, dead‑letter queues, delay and ordered messages, transactional flows, high‑availability setups, load balancing, partitioning, leader election, offset handling, and practical solutions to common messaging challenges.
This article surveys common delayed‑message solutions in distributed asynchronous messaging, evaluates implementations based on external storage, databases, RocksDB, Redis, and open‑source MQs like RocketMQ, Pulsar and QMQ, and discusses their advantages, drawbacks, and practical design considerations.
This article provides an in‑depth technical guide to three major message‑queue systems—RabbitMQ, RocketMQ, and Kafka—covering their core components, exchange and routing mechanisms, durability features, consumer acknowledgment, dead‑letter handling, load balancing, and practical strategies for reliability, ordering, and scaling in distributed backend applications.
This article explains the underlying principles of Sentinel's cluster rate‑limiting feature, compares embedded and independent deployment modes, and provides a step‑by‑step Java/SpringBoot implementation with Apollo dynamic configuration and sample test code.
vivo’s Internet Middleware Team replaced its RabbitMQ service with a RocketMQ‑based platform, building an AMQP‑Proxy gateway and metadata layer to enable seamless, zero‑downtime migration while achieving over 100,000 TPS, billion‑message capacity, 50% resource savings, and advanced features such as transactions, ordered and delayed messaging, and tracing.
This article provides an extensive overview of microservice architecture, tracing its evolution from monolithic to distributed services, detailing the advantages and drawbacks of each stage, and presenting a practical technology stack—including service discovery, API gateways, containerization, orchestration, logging, monitoring, and CI/CD—to help engineers design, implement, and operate modern cloud‑native systems.
This comprehensive guide explains Elasticsearch’s role in searching structured and unstructured data, covers Lucene’s inverted index, details cluster components, shard and replica mechanics, mapping types, installation steps, indexing workflow, storage strategies, and practical performance optimizations for real‑world deployments.
This article explains the fundamentals of service governance in distributed systems, covering RPC communication, common anomalies, replica types, design principles, performance metrics, and the architecture and workflow of RPC frameworks, while also detailing key practices such as registration, discovery, load balancing, traffic shaping, version compatibility, circuit breaking, degradation, and rate limiting.
This article examines four strategies for handling session data in distributed backend systems—session replication, client-side storage, hash-based load balancing, and centralized storage with Redis—detailing their principles, advantages, disadvantages, and practical recommendations for different deployment scales.
This article explains the differences between monolithic and microservice architectures, outlines their advantages and disadvantages, and then dives into essential components such as service gateways, service registration and discovery, and distributed configuration centers, providing practical guidance for building scalable backend systems.
Chaos engineering, a discipline for experimenting on distributed systems, helps teams identify hidden weaknesses, improve high‑availability, and build confidence in production by defining stable states, injecting realistic failures, and measuring impact through observability metrics, with practical steps, tool choices, maturity stages, and evaluation methods.
This article provides a detailed introduction to etcd, covering its purpose as a highly available distributed key‑value store, core concepts like Raft consensus, key features, common use cases such as service discovery and configuration management, step‑by‑step installation for single‑node and cluster deployments, and essential etcdctl commands for managing data and cluster members.
This article provides a comprehensive overview of Kafka’s core components—including brokers, topics, partitions, producers, and consumers—explaining their roles, log structures, replication mechanisms, and the system’s architecture, supplemented with practical visual illustrative diagrams.
The article analyzes the shortcomings of traditional ELK and distributed tracing for complex business systems, proposes a visualized full‑chain log tracing solution that organizes and dynamically links logs by business chain, and demonstrates its implementation and performance gains at Meituan’s content platform.
This article details a low‑impact migration from MyCat to DBLE, covering required parameter adjustments, sharding function replacement, data node handling, and business‑SQL compatibility issues, while providing concrete configuration examples and practical recommendations for a smooth transition.
The article explains why simply adding hardware cannot keep up with growing database workloads and presents a three‑step approach—business isolation, horizontal sharding, and advanced reporting using PolarDB‑X—to achieve high concurrency, elastic capacity, and efficient caching for modern high‑traffic applications.
This article explains how to design and implement a precise, persistent, and retry‑capable delay queue using Redis ordered sets and Lua scripts, provides a complete Go library with usage examples, and discusses the underlying data structures, atomic operations, and garbage‑collection mechanisms.
This article explains common load‑balancing strategies—including round‑robin, random, weighted, smooth weighted round‑robin, consistent hashing, least‑active, and fastest‑response algorithms—provides Java implementations with code samples, discusses their advantages, disadvantages, and suitable scenarios, and offers practical guidance for choosing the right method in distributed systems.
This article examines the historical development, distributed architecture, and high‑concurrency challenges of China’s 12306 railway ticketing platform, tracing its origins from early Unix‑based systems to modern multi‑layered backend solutions that support hundreds of millions of users during peak travel periods.
This article explains cloud‑native observability, defines its three pillars—metrics, tracing, and logging—details the OpenTelemetry tracing data model and Span structure, reviews industry implementations such as Jaeger and Alibaba Eagle Eye, and shares practical challenges and solutions from real‑world production use.
This article presents a comprehensive overview of a hand‑crafted RPC framework built with Spring Boot, Netty, and Zookeeper, covering its core concepts, architecture, custom message protocol, load‑balancing strategies, serialization choices, starter design, deployment steps, and testing procedures.
The article explains the benefits, costs, and various update, penetration, no‑hole, avalanche, and hot‑key optimization techniques for caching in high‑concurrency distributed systems, providing practical guidance on choosing appropriate strategies based on consistency requirements and system load.
This article explains the principles behind popular load‑balancing strategies such as round‑robin, weighted round‑robin, smooth weighted round‑robin, consistent hashing, least‑active, and best‑response, and provides clear Java code examples for each method.
This article examines Meituan's five‑year evolution of instant logistics, detailing the distributed, high‑concurrency architecture, AI‑driven optimization, scalability techniques, fault‑tolerance mechanisms, and future challenges faced by its real‑time delivery platform.
This article explains how Meituan's instant logistics platform evolved from a monolithic system to a highly available, scalable, and AI‑enhanced distributed architecture, detailing the technical challenges, architectural upgrades, fault‑tolerance strategies, and future scalability concerns.
This article details Meituan's instant logistics platform architecture, covering its background, distributed system design, high‑availability deployment, AI‑driven optimization, and future challenges, while sharing practical solutions for scalability, fault tolerance, and operational efficiency in a high‑concurrency environment.
This article reviews common load balancing strategies—including round-robin, random, weighted, smooth weighted round-robin, consistent hashing, least-active and optimal-response algorithms—explains their advantages and drawbacks, and provides complete Java code examples for each method.
This article explains the fundamentals of message middleware, covering its architecture, components, communication models, advantages, common protocols, major open‑source solutions, detailed feature comparisons, and practical guidance for selecting the most suitable message queue based on system scale, business needs, performance, high availability, and operational considerations.
This article explains common load‑balancing strategies—including basic round‑robin, random, weighted and smooth weighted algorithms, as well as consistent hashing, least‑active and optimal‑response methods—provides Java implementations for each, discusses their advantages, disadvantages, and suitable scenarios, and concludes with practical recommendations.
This article explains the concept of interface idempotency, demonstrates how to achieve idempotent update operations using version control and token mechanisms, and provides a comprehensive guide to distributed rate limiting—including time‑window and resource‑based dimensions, token‑bucket and leaky‑bucket algorithms, and practical implementations with Guava RateLimiter, Nginx, and Redis‑Lua scripts.
RocketMQ consumers initialize by validating configuration, creating subscription data, establishing a client instance, loading offsets, and starting services before registering with the client factory, after which they can receive messages via push (default low‑latency) or pull modes, support ordered processing, and handle acknowledgments with configurable retries and dead‑letter handling.
This article explains how to design and implement a high‑concurrency system for e‑commerce scenarios, covering server architecture, load balancing, database clustering, caching strategies, concurrency testing tools, message‑queue based async processing, first‑level cache, static data handling, layering, distribution, redundancy and automation, with concrete examples such as user sign‑in, order queries and red‑packet distribution.
Network timeouts do not necessarily indicate server‑side failure; handling them with appropriate retry strategies, exponential backoff, and idempotent APIs—combined with mechanisms such as distributed locks and atomic transactions—helps maintain system stability while avoiding duplicate operations and resource exhaustion.
Bilibili’s load‑testing platform evolved from ad‑hoc JMeter scripts to a fully automated, self‑service system (Platform 2.0) that uses a custom load client, adaptive scheduling, and flexible scenario modes—including traffic replay and data‑isolated testing—to efficiently stress‑test over a hundred microservices for large‑scale events, with further integration and circuit‑breaker enhancements planned.
Meta engineers describe the challenges of cache invalidation and consistency in large‑scale distributed systems, explain why stale caches are problematic, present their Polaris observability service and consistency‑tracking techniques, and detail how they raised TAO’s cache consistency from six‑nines to ten‑nines.
This article examines the trade‑offs of common distributed ID generators—UUID, timestamp‑based schemes, Snowflake, and NanoID—highlighting their size, ordering, performance, and security implications to help developers select the most suitable approach for their systems.
This article details Ctrip's large‑scale, real‑time log monitoring solution, covering the overall monitoring architecture, the Clog log system, the CAT tracing platform, and the TSDB metric store, and explains design choices such as write‑heavy indexing, segment‑based storage, and migration to ClickHouse for high‑cardinality data.
This article explores the importance of rate limiting in distributed micro‑service architectures, explains four core algorithms—fixed window, sliding window, leaky bucket, and token bucket—and details a practical, Redis‑backed multi‑layer throttling solution for a voice‑bot platform, including trade‑offs and implementation tips.
This article explains how Eureka’s heartbeat mechanism works in microservice architectures, covering who sends the heartbeat, the scheduling interval, the request implementation, server-side handling, and the effect on service registration, using code examples and analogies to historical tribute systems.
This issue curates five technical articles covering Gaode's full‑chain stress testing platform, a distributed web UI automation system, a comprehensive WWDC22 session review, ShouQianBa's efficient automated testing practices, and Autohome's AutoDiff traffic replay for integrated testing.
This article explains the OSI seven‑layer model, compares RPC and HTTP services, details RPC architecture, synchronous vs asynchronous calls, and reviews popular RPC frameworks, helping developers choose the right approach for enterprise applications.
This article explores common cache challenges such as centralized expiration, cache penetration, avalanche, hot keys, large keys, data consistency, and concurrent pre‑warming, offering practical design patterns and mitigation strategies to build robust, high‑performance backend systems.
This article details Qunar Travel's mature chaos engineering platform built on chaosblade, covering value analysis, system architecture, shutdown and dependency drills, automated closed‑loop testing, attack‑defense exercises, and the measurable reliability improvements achieved across thousands of services.
This article explores the core goals of microservice architecture, outlines key design principles such as scalability and resilience, and presents a comprehensive catalog of decomposition, integration, cross‑cutting concern, and observability patterns—including API gateway, Strangler, Bulkhead, Saga, and CQRS—to guide developers in building robust, maintainable backend systems.
This article explains the Paxos consensus algorithm—its origins, core concepts, roles of proposers, acceptors and learners, safety and liveness constraints, the two-phase protocol, proposal generation, and practical variations—showing why Paxos remains a foundational solution for fault‑tolerant distributed systems.
This article explains the split‑brain phenomenon in distributed systems, illustrates how it can occur in Zookeeper clusters, and details Zookeeper's quorum‑based solutions—including majority voting, odd‑node deployment, and additional safeguards—to avoid split‑brain and ensure reliable leader election.
The article analyzes why modern software systems repeatedly collapse—due to growing business complexity, unpredictable changes, and architectural decay—and proposes principles such as decentralization, integration, and diversity, along with practical strategies like service mesh and eBPF, to design more sustainable, observable, and self‑evolving architectures.
This article explores Redis cache mechanisms, various caching patterns such as Cache‑Aside, Read‑Through, Write‑Through and Write‑Behind, analyzes consistency challenges in write operations, and presents solutions like delayed double deletion, retry mechanisms, and binlog‑based asynchronous cache invalidation to achieve eventual consistency.
This article visually breaks down Kafka’s fundamental concepts—including topics, partitions, producers, consumers, consumer groups, and cluster architecture—so readers can grasp how messages flow, are stored, and achieve load balancing and ordering within a distributed streaming platform.
This article examines ClickHouse's join mechanisms, detailing the limitations of standard joins, the advantages of Global joins, and optimization strategies such as hash and merge joins, subquery filtering, and memory considerations, illustrated with SQL examples and source‑code analysis.
This article details JD Retail's OLAP platform construction, covering control plane design, architecture, business and operation management, real‑time data updates, materialized view usage, join optimizations, high‑concurrency and high‑throughput scenarios, and promotional preparation strategies, illustrated with diagrams and performance metrics.
This article chronicles ByteDance's journey from early MySQL‑based databases to a sophisticated, cloud‑native, distributed database platform that now supports over 10 EB of storage, detailing the challenges, architectural milestones, and future directions of its database infrastructure.
This article introduces and compares four widely used open-source distributed message-queue systems—Kafka, ActiveMQ, RabbitMQ, and RocketMQ—detailing their architectures, deployment requirements, features, high-availability solutions, advantages, disadvantages, and provides guidance for selecting the appropriate middleware in large-scale applications.
This article explains how RocketMQ tackles registration latency, synchronous coupling, and traffic spikes by introducing an intermediate queue, designing a persistent high‑availability broker, leveraging Linux page cache and memory‑mapped files, and employing topics, tags, and sharding to enable efficient asynchronous processing and scalable consumption.
The article explains how RocketMQ tackles growing business traffic by introducing an asynchronous broker layer, using commitlog and consumeQueue files, page‑cache, mmap, topic/tag routing, and a nameserver to achieve high‑throughput, low‑latency, and highly available message delivery.
