This article enumerates essential operational metrics—such as CPU, memory, disk and network I/O, response time, throughput, error rates, availability, MTBF/MTTR, security logs, and capacity‑planning indicators—explaining their meanings and recommended target values to help engineers comprehensively monitor system performance, stability, and efficiency.
The article outlines why monitoring key IT operations metrics is vital for performance, reliability, and cost control, then details 17 common metrics—including availability, failure rate, MTTR, MTBF, response time, throughput, error rate, capacity utilization, latency, data integrity, success rates, waiting time, backup success, recovery time, security patch time, server and network bandwidth utilization—providing definitions, calculation formulas, typical reference values, and applicable scenarios.
This article explains the differences between reliability, availability and maintainability, introduces MTTF, MTTR and MTBF, shows how to compute availability correctly, and walks through series, parallel and mixed system calculations with concrete examples from a medical‑platform architecture.
The article explains why vague quality goals like "fast, stable, secure" are insufficient for architecture design and shows how to turn them into measurable quality‑attribute scenarios using six fixed elements—stimulus source, stimulus, environment, artifact, response, and response metric—illustrated with concrete healthcare system examples.
The article explains the CAP theorem and BASE model by walking through a concrete scenario of two data centers losing network connectivity, showing how architects must choose between consistency and availability and illustrating typical CP and AP use cases.
During peak events like Alibaba's Double‑11 and WeChat's red‑packet frenzy, distributed systems must trade strict consistency for availability; this article explains the CAP theorem, introduces the BASE model, compares CP and AP designs, and provides real‑world case studies and selection guidelines.
Every delay you notice in WeChat messages, flash‑sale pages, or bank transfers stems from the same underlying distributed‑system trade‑off, and this article explains the CAP theorem, its three guarantees, common misconceptions, and how CP versus AP architectures shape real‑world services.
This article explains the CAP theorem and its practical extension BASE, describing their core concepts, trade‑off combinations, typical components such as Zookeeper, Eureka, and Nacos, and engineering techniques like asynchronous replication, Saga, and idempotent design for building highly available distributed systems.
At the 2024 GOPS Global Operations Conference in Shanghai, Alipay’s monitoring lead Tang Liang presented the challenges, architecture, risk‑prevention practices, and implementation details of the company’s full‑ecosystem availability monitoring system, highlighting its role in DevOps, SRE, and AIOps initiatives.
This article examines DeepSeek’s frequent service interruptions, contrasts the inherent reliability challenges of AIGC products with traditional internet applications, and proposes adopting proven isolation, rate‑limiting, and elastic‑scaling techniques to improve AI service availability and user experience.
An in‑depth guide explains the CAP theorem’s three pillars—Consistency, Availability, Partition Tolerance—illustrates why only two can be achieved simultaneously, and shows real‑world trade‑offs across e‑commerce, finance, and social platforms, while introducing the complementary BASE model for practical system design.
This article explains why operations metrics are vital for businesses, describes how tracking availability, failure rate, MTTR, MTBF, response time, throughput, error rate, capacity utilization, latency, data integrity, backup success, recovery time, security patch time, server and network utilization can improve reliability, reduce costs, and boost competitiveness.
This article explores the evolution of software complexity and presents a comprehensive backend development methodology for building high‑performance, high‑concurrency, and highly available systems, covering performance optimization, read/write strategies, scaling techniques, fault isolation, and deployment practices with real‑world examples.
This guide outlines key operational metrics—availability, failure rate, MTTR, MTBF, response time, throughput, error rate, capacity utilization, latency, data integrity, and more—explaining their calculations, typical benchmark values, and practical application areas to help organizations monitor and improve IT performance.
This article explains the core concepts of the BASE theory—Basically Available, Soft State, and Eventual Consistency—illustrating how distributed systems trade off consistency for higher availability and performance, and provides practical examples and typical application scenarios.
By integrating the proactive, static risk assessment of Failure Mode and Effects Analysis with the dynamic fault‑injection validation of chaos engineering, the article demonstrates how cloud‑native architectures—illustrated through a Tencent‑based e‑commerce case—can systematically identify, quantify, and mitigate availability risks, leading to continuous, measurable resilience improvements.
This article explores how service availability is quantified, explains the impact of MTBF and MTTR on reliability, and presents concrete operational practices—including redundancy, traffic control, and change‑management techniques—to move systems from basic uptime to true high‑availability levels.
In today’s competitive business landscape, tracking key operations metrics—such as availability, failure rate, MTTR, MTBF, response time, throughput, error rate, and various utilization and data integrity measures—helps organizations monitor performance, reduce costs, ensure reliability, and maintain regulatory compliance.
The article outlines a three‑stage evolution of operations—from rapid incident response to proactive fault‑injection—while offering practical guidance on improving availability, visualizing changes, and aligning technical metrics with business value to elevate the role of operations engineers.
The article details Tencent's Public Gateway (TGW) architecture, its forwarding and control planes, and presents two real‑world extreme failure cases— a NIC batch bug and a special IPv6 packet causing core dumps—along with the multi‑level disaster‑recovery design and mitigation strategies employed to ensure high availability.
Learn how the CAP theorem defines the trade‑offs among consistency, availability, and partition tolerance, and see how different NoSQL databases such as MongoDB (CP) and Cassandra (AP) align with these principles to guide your choice of storage for cloud‑native, distributed applications.
Understanding the differences between system availability, reliability, and stability is essential for building resilient services; this guide explains each concept, illustrates their distinctions with examples, and outlines practical strategies such as rate limiting, anti‑scraping, timeout settings, system inspections, and fault post‑mortems to reduce failures and downtime.
The article details how to design and implement a cloud‑native payment fund account system on Tencent Cloud, covering account definitions, fund flow and multiple account types, TDSQL storage, separated fund and account services, robust security, distributed transactions, auditing, reconciliation, and high‑availability measures for high‑concurrency merchant payments.
The article explores why building system stability is crucial by linking entropy theory to software reliability, introduces the availability formula, discusses common pitfalls and industry practices, and proposes a three‑stage governance framework—prevention, mitigation, and post‑mortem—to systematically improve operational resilience.
This article traces the evolution of Tencent's payment platform from its early PC‑centric design through three mobile payment phases, detailing architectural generations, availability measures, multi‑active strategies, and cloud‑native innovations that enable massive, reliable transaction processing.
The article examines the March 29 Vipshop data‑center outage that caused over a billion‑yuan loss, explains the cooling‑system failure that triggered a 12‑hour P0 incident, discusses its impact on Tencent services, and analyzes why high‑concurrency crashes remain common, offering availability tier insights and mitigation strategies.
This article explains the CAP theorem and its trade‑offs, introduces the BASE model as a practical compromise, and demonstrates how Zookeeper implements a CP registration center while Eureka adopts an AP approach, illustrating the impact on consistency, availability, and partition tolerance in distributed systems.
The article explains ten key non‑functional quality attributes of software architecture—such as scalability, availability, consistency, resilience, usability, observability, security, persistence, agility, and maintainability—describing their meanings, typical implementation techniques, and why selecting the right attributes is crucial for any system.
This article clarifies the often‑confused concepts of system stability, high availability, reliability and availability, explains their metrics such as MTBF, MTTR and MTTF, and shows how they interrelate to guide engineers in building resilient services.
The article reviews traditional availability metrics such as Success‑Ratio, Error‑Budget, MTTR/MTTF, SLA/SLO, and highlights their limitations, then introduces Google’s User‑Uptime and Windowed User‑Uptime metrics, explains their definitions, challenges, experimental results, and why they provide a more user‑centric view of service reliability.
The article reviews classic availability metrics such as Success‑Ratio, Incident‑Ratio, MTTR/MTTF, Error‑Budget, and SLA/SLO, then introduces User‑Uptime—a per‑user success time proportion that ignores long idle periods—and its windowed variant, showing how it complements existing indicators for more user‑centric reliability insight.
The article explains the CAP theorem’s three properties—consistency, availability, and partition tolerance—its implications for distributed systems, highlights its limitations, introduces the PACELC extension that adds latency versus consistency trade‑offs when no partition exists, and provides real‑world database examples.
The article revisits the CAP theorem, explains linearizable consistency and strict availability, demonstrates with concrete data‑center examples why the CP/AP split is often misleading, and argues that latency concerns and broader failure modes make the binary classification of distributed systems impractical.
The article explains the CAP theorem’s origin, formal proof, detailed definitions of consistency, availability and partition tolerance, its impact on relational and NoSQL databases, various consistency models, and practical trade‑offs for designing distributed data systems.
This article explains the origin of the CAP theorem, its formal definitions of consistency, availability, and partition tolerance, examines how it applies to relational and NoSQL databases, discusses trade‑offs and practical strategies for balancing these properties in distributed system design.
The article outlines four essential cloud‑application design dimensions—scalability, availability, manageability, and feasibility—providing discussion points and questions for each to guide stakeholders toward robust, cost‑effective, and secure cloud solutions through comprehensive evaluation of capacity, platform constraints, load handling, SLA commitments, disaster recovery, performance tuning, and security considerations.
This article examines Baidu Search's massive micro‑service architecture and reveals the detailed observability, tracing, and metrics techniques—Kepler 1.0, Kepler 2.0, and Prometheus integration—that enable five‑nine‑plus availability, full‑query debugging, and efficient capacity management.
The article analyzes the limitations of using ZooKeeper for service discovery, compares it with Alibaba's ConfigServer, discusses CAP trade‑offs, scalability, health‑check design, disaster recovery, and provides practical guidance on when to choose alternative registration solutions.
The article explains the origins of the CAP theorem, defines its three core properties, illustrates why a distributed system can satisfy at most two of them, discusses practical workarounds such as eventual consistency, and provides references for further study.
This article explores how the procurement system at Yanxuan adapts to shifting commercial environments and unexpected events by defining macro-level logic, designing blueprints, implementing the system, and continuously evolving it, emphasizing the importance of top‑down architecture, scalability, availability, accuracy, and the transition toward automation and intelligence.
This article uses the story of Tai Chi from the novel *The Heaven Sword and Dragon Saber* to explain the CAP theorem, ACID properties, BASE theory, and two‑phase commit in distributed systems, illustrating how consistency, availability, and partition tolerance correspond to the hard and soft aspects of Tai Chi.
The article compares Eureka and Zookeeper as service registry solutions, explaining how Eureka follows an AP model with high availability and eventual consistency, while Zookeeper adopts a CP model prioritizing strong consistency, and discusses their suitable scenarios, limitations, and design considerations for distributed systems.
Quorum, the core principle behind majority read/write and Paxos, can be defined in various ways—including weighted, hierarchical, and non‑majority quorums—to trade off system availability, latency, and fault tolerance, with examples illustrating how different quorum designs affect performance in distributed storage and coordination services.
The article presents iQIYI’s service‑maturity model for micro‑services, outlines how scores across development, deployment and operation stages reveal common deficiencies such as code style, testing, gray‑release and alert handling, and recommends concrete optimization practices—including unified coding standards, automated testing, robust rollback, circuit‑breaking, monitoring, and emergency procedures—to raise services to mature, high‑scoring levels.
This article uses a relatable memory‑service scenario to illustrate the CAP theorem, explaining how consistency, availability, and partition tolerance cannot all be achieved simultaneously in distributed systems and exploring practical trade‑offs through successive design attempts.
The article outlines a three‑stage ops maturity model—firefighting, fire prevention, and arson—explains how proactive fault‑injection drills, continuous availability improvements, and aligning technical metrics with business value can transform operations from reactive responders into strategic value creators.
Distributed locks ensure exclusive access to shared resources across multiple machines, and this article explains their evolution from single-machine locks, classifies system designs, and details Alibaba Cloud Storage’s practical implementation, covering strict mutual exclusion, availability, and lock-switching efficiency with real-world examples.
The article explains why Zookeeper’s CP consistency model and performance characteristics make it unsuitable for high‑availability service registries, advocating an AP‑oriented approach for service discovery in large‑scale systems.
Tencent Video’s front‑end Node.js services achieve massive concurrency stability through a layered architecture that combines GSLB‑directed CDN, TGW, Nginx, and clustered workers, reinforced by process guardians, three‑tier disaster‑recovery fallbacks, multi‑level caching with lock mechanisms, and comprehensive logging and alerting.
This article explains how to build high‑performance, highly‑available, scalable, extensible, and secure recommendation services by outlining background concepts, defining quality criteria, discussing design challenges, and presenting concrete architectural principles and practical strategies.
This guide outlines six practical strategies—rollback testing, cautious destructive actions, clear command prompts, verified backups, careful handovers, and proactive monitoring—to help operations teams minimize outages and maintain high system availability.
This article details Youzan's transition from Hive to SparkSQL, covering platform architecture, usability and performance enhancements, migration strategies, automated engine selection, and future plans that together reduced resource consumption by up to 67% while handling thousands of daily jobs.
This article explains the CAP theorem, its implications for distributed systems, details the Paxos consensus algorithm and its availability optimizations, and introduces ZooKeeper's ZAB protocol, showing how they together address consistency‑availability trade‑offs in real‑world deployments.
The article explains cloud computing fundamentals, details Service Level Agreements (SLAs) and their metrics, compares the availability and compensation policies of major Chinese cloud providers, and concludes with a brief DevOps recruitment notice, highlighting both technical insights and industry context.
The article outlines Meituan Finance’s front‑end availability challenges in its million‑order payment service and presents a disciplined, end‑to‑end approach—standardized release processes, simple fallback designs, automated testing, robust monitoring, and regular fault‑drill simulations—to ensure stable user experiences across diverse client environments.
The article explains the CAP theorem, its three properties—consistency, availability, and partition tolerance—how they influence database design, compares CA, CP, and AP classifications with examples, and discusses related concepts such as ACID, BASE, and practical design choices for distributed systems.
This article compares the architectures of Kafka and RocketMQ, examines their availability and reliability mechanisms, evaluates their strengths and weaknesses, and proposes hybrid designs and simplified MQ architectures for building highly available and reliable messaging systems.
This guide outlines the five essential architecture metrics—performance, availability, scalability, extensibility, and security—detailing practical optimization techniques for front‑end resources, server‑side caching, database tuning, load balancing, and security layers to build resilient, high‑performing web systems.
This article examines the growing challenges of system availability in modern operations, explains the concept of availability and the N‑nine metric, introduces Business Continuity Management and Google SRE approaches, and provides concrete technical and managerial methods—including architecture standardization, scaling strategies, tooling, emergency drills, and incident‑centralized management—to improve operational reliability.
This article demystifies reliability by clarifying the difference between durability and availability, exposing common misconceptions about MTBF, analyzing real‑world disk failure data, and presenting a practical formula for calculating the health probability of distributed storage systems.
The 2017 China Cloud Computing Evaluation Report by Tingyun analyzes major cloud providers across user experience, performance, availability, and ease of use, using a 132‑item questionnaire and verification tests to deliver regional performance scores and comprehensive provider rankings.
The article outlines Meituan‑Dianping’s approach to front‑end service availability for its checkout system, defining availability across code, static resources, and network links, measuring failure duration, identifying typical bugs, and implementing a three‑stage assurance strategy using people processes, engineering tools, lightweight technology choices, and concrete practices such as TypeScript adoption, automated testing, health‑checks, DNS protection, and post‑incident monitoring.
This article explains website availability metrics, fault‑weight scoring, layered high‑availability architecture, session management strategies, reusable service design, data redundancy, quality assurance processes, and monitoring practices essential for maintaining reliable large‑scale web systems.
The article explains the fundamental architectural factors of large‑scale web systems—performance, availability, scalability, extensibility, and security—detailing practical optimization techniques, measurement metrics, and design principles that guide robust software architecture decisions.
The article recaps a technical talk by Qunar’s accommodation trading system lead, covering event‑driven design, flexible document schemas, and reliability techniques such as circuit breaking, gray releases, and automated testing to improve scalability and maintainability of backend transaction platforms.
This article examines the architectures of Kafka and RocketMQ, analyzes their availability and reliability mechanisms, evaluates their strengths and weaknesses, and proposes a hybrid MQ design that combines the benefits of both systems while simplifying dependencies and improving fault tolerance.
This article explains the Trusted Cloud host rating framework, detailing its star‑based levels, evaluation criteria such as availability, security and disaster recovery, and how enterprises can use the standards to select the most suitable cloud host provider.
This article explores how Google Site Reliability Engineers manage service reliability by balancing risk, cost, and business goals, using metrics like unplanned downtime, availability formulas, and risk tolerance to set realistic SLOs for both consumer and infrastructure services.
The article explains how reliability of storage, servers, and distributed systems is assessed using standards, models like MTBF/MTTR, RAS features, CAP/BASE theories, and end‑to‑end solutions, emphasizing the gap between theoretical metrics and real‑world operational data.
This article explains the CAP theorem—its three core properties, why a distributed system can satisfy at most two of them, the formal proof, practical strategies for handling partitions, and how modern databases and architectures attempt to work around its constraints.
This article outlines a comprehensive framework for assessing IT operations maturity by defining four core dimensions—availability, cost, efficiency, and technological advancement—along with quantitative metrics, scoring formulas, and practical methods for data collection and continuous performance improvement.
The article examines system reliability concepts such as availability, MTBF, MTTR, and outage classifications, analyzes the Alipay service interruption, discusses various redundancy and failover strategies, and explores YunOS reliability testing and design practices to improve overall system robustness.
The article explores the essence of information security by comparing it to safeguarding personal money, detailing the four fundamental attributes—confidentiality, integrity, availability, and controllability—and illustrating how different conditions shape security needs, from personal to enterprise contexts.
This article explains how to objectively assess cloud service reliability by examining three key aspects—availability, access control, and disaster recovery—and provides practical strategies such as redundancy design, gradual deployment, automation, and robust backup to improve overall cloud service trustworthiness.
This article examines key considerations for building highly scalable and available systems on Azure, outlining four architectural dimensions—scalability, availability, manageability, and feasibility—while discussing patterns, anti‑patterns, measurable resources, queue‑based load balancing, authentication services, and common pitfalls such as configuration errors and SQL injection.
