This tutorial walks you through setting up a three‑node Hadoop 2.9.2 cluster on CentOS 7.5, covering environment preparation, password‑less SSH, user creation, JDK installation, Hadoop extraction, configuration file edits, directory setup, ownership changes, service startup, and verification via web UIs.
The article explains Spark’s executor memory layout under the UnifiedMemoryManager, detailing on‑heap and off‑heap divisions, the four memory regions, default fraction settings, how storage and execution memory share space, and provides heuristics and tuning tips for avoiding OOM and optimizing performance.
This article explains how to tune Hive on Spark by adjusting driver, executor, and Hive configuration parameters—including CPU cores, memory allocations, dynamic allocation, and join thresholds—to achieve optimal performance when running on YARN.
This article details Ctrip's cross‑datacenter Hadoop architecture, covering the evolution of its Hadoop platform, the challenges of multi‑site bandwidth and latency, design choices between multi‑cluster and single‑cluster solutions, and the concrete HDFS, YARN, balancer, migration, monitoring, and throttling implementations that enable transparent, consistent, and efficient multi‑datacenter operations.
The article describes how to troubleshoot and fix Oozie shell‑action failures when submitting Flink jobs on a CDH 6.3 cluster with Kerberos, detailing environment‑variable conflicts, error messages, and the final solution using a clean environment and custom FLINK_CONF_DIR settings.
The article explains Spark's default static resource allocation, analyzes the limitations of its Dynamic Resource Allocation (DRA) for streaming workloads, describes the internal Spark components and code paths involved, and proposes concrete design and configuration recommendations for implementing more responsive executor scaling.
This article explains the role of YARN's Scheduler, compares FIFO, Capacity, and Fair schedulers, details their configurations—including XML snippets for Capacity and Fair schedulers, queue hierarchy, preemption settings, and provides practical guidance for resource allocation in Hadoop clusters.
The article details the technical evolution of 58.com’s real-time computing platform—from Storm and Spark Streaming to a Flink‑based one‑stop solution called Wstream—covering use cases, architecture, stability measures, migration from Storm, operational diagnostics, and future development plans.
This article presents a comprehensive analysis of Meituan's Hadoop YARN fair scheduler, detailing its architecture, resource abstractions, scheduling workflow, performance bottlenecks, fine‑grained metrics, and a series of optimization techniques—including sorting improvements, job‑skip reduction, parallel queue sorting, and robust rollout strategies—to achieve high‑throughput, low‑latency scheduling for large‑scale offline, streaming, and machine‑learning workloads.
Jiaoxiao Fang’s talk details Kuaishou’s YARN deployment, covering its architecture, support for offline, real‑time and ML workloads, and recent enhancements such as event‑handling stability, refined preemption, high‑throughput parallel scheduling, shuffle‑caching for small I/O, plus plans for job protection and multi‑cluster resource utilization.
The cloud+ community and Kuaishou hosted a big‑data technology salon where experts detailed the evolution, architecture, and practical deployments of Spark‑based cloud data warehouses, ElasticSearch, Yarn, and Flink, highlighting trends, optimization techniques, and future directions for enterprise data analytics.
This article details Meituan's experience optimizing the Hadoop YARN fair scheduler, covering background challenges, architectural components, resource abstractions, scheduling flow, performance metrics, a series of code‑level optimizations, stability strategies for production rollout, and future directions for large‑scale cluster scheduling.
This article describes how the 58 real‑time computing platform optimized the Flink‑Storm beta tool and implemented large‑scale, smooth migration of Storm jobs to Flink, covering background, architecture, platform‑level enhancements, YARN runtime support, deployment, and user‑side integration.
Meituan improved its custom Hadoop YARN Fair Scheduler by pre‑computing resource usage, filtering zero‑demand jobs, and parallelizing queue sorting, which reduced sorting time from 30 s to 5 s per minute, boosted container‑per‑second throughput to 50 k, enabled live roll‑backs, and prepared the system for clusters up to 10 k nodes and future scaling to hundreds of thousands.
This article provides a comprehensive, language‑agnostic tutorial on building a highly available Hadoop cluster, covering HDFS and YARN HA architectures, QJM shared storage, required components, configuration files, installation commands, startup procedures, verification steps, and troubleshooting references.
This article provides a step‑by‑step tutorial on configuring Hadoop high availability, covering HDFS HA architecture, Quorum Journal Manager synchronization, NameNode failover, YARN HA, required pre‑conditions, cluster planning, configuration files, service startup, and verification procedures.
This article explores Hadoop’s role in the big‑data era, detailing its architecture, core components such as HDFS, YARN, MapReduce, Ozone and Submarine, the challenges of trillion‑scale data, and why its scalability, cost efficiency, and a mature ecosystem give it a competitive edge.
This weekly bulletin summarizes four Hadoop knowledge points—compression formats, MapReduce join techniques, Hive installation, and YARN Capacity Scheduler—while also sharing personal updates about a PhD graduation, the upcoming May Day holiday, and a request for likes and shares.
The article provides a comprehensive overview of Hadoop's Capacity Scheduler, describing its resource‑allocation features, configurable XML parameters, queue access controls, dynamic configuration updates, and the internal workflow of application initialization and resource scheduling within YARN.
This article explains why YARN was created to overcome the limitations of MapReduce 1.x, describes its architecture—including ResourceManager, NodeManager, ApplicationMaster, Container, and Client—and outlines the step‑by‑step execution flow that enables multiple computation frameworks to run on Hadoop.
This guide explains how to set up an Apache Flink cluster on CentOS 7 using three deployment methods—Local, Standalone, and Flink on YARN/Kubernetes—including host configuration, SSH setup, package distribution, configuration file editing, cluster start/stop commands, YARN resource manager concepts, session commands, job submission, fault‑tolerance settings, and log inspection.
The article describes a log‑parsing feature for Youzan’s big‑data offline platform that captures runtime logs from Hive, Spark, DataX, MapReduce and HBase jobs, categorizes scheduling types, extracts metrics such as read/write bytes, shuffle volume and GC time, and processes them in real time via a Filebeat‑Logstash‑Kafka‑Spark‑Streaming pipeline storing results in Redis for monitoring, optimization and resource‑usage ranking.
This article walks through Youzan's real‑time platform architecture, explains why Flink was chosen over Spark Structured Streaming, details practical challenges such as container over‑provisioning and monitoring overhead, shares solutions for Spring integration and async caching, and outlines future directions for SQL‑based streaming and scheduler improvements.
This tutorial explains how to deploy Apache Flink on a Hadoop YARN cluster, covering both YARN session mode and direct job submission, and demonstrates running the built‑in WordCount example with command‑line options for input, output, and resource configuration.
This article describes how JD built a multi‑regional, ten‑thousand‑node Hadoop ecosystem, unified resource management with YARN, introduced a three‑level Router scheduling layer, optimized performance, and integrated deep‑learning frameworks to achieve high availability, cost efficiency, and scalable big‑data processing.
This article introduces Hadoop’s origins and core challenges, then presents thirteen essential open‑source tools spanning resource scheduling, real‑time query engines, and additional processing frameworks, detailing each project's purpose, key features, and repository locations to help practitioners choose the right component for big‑data workloads.
Despite Gartner's 2017 claim that Hadoop is nearing the end of its production maturity, a series of interviews with Chinese big‑data experts reveal that Hadoop's ecosystem remains robust, with core components like HDFS, YARN, Spark, and HBase continuing to dominate the market.
This article explains Spark's role in the DataMagic platform, outlines four practical steps to quickly master Spark, details key configuration and parallelism settings, shows how to modify Spark code, and provides operational tips for cluster management and job troubleshooting.
This article explains Hadoop's core components, the MapReduce programming model, the detailed shuffle and merge processes, and how YARN replaces the classic JobTracker/TaskTracker architecture to improve scalability and resource utilization in large‑scale data processing clusters.
This article explains Hadoop's distributed storage and processing framework, details the MapReduce programming model, describes the classic JobTracker/TaskTracker architecture, outlines the shuffle and combine phases, and introduces YARN as a scalable replacement with its ResourceManager, ApplicationMaster, and NodeManager components.
This guide explains how DataMagic leverages Spark on YARN for fast, scalable offline analytics—covering Spark’s core role, four steps to master its terminology, configurations, parallelism, and code modification, plus practical deployment scripts, dynamic resource tuning, MongoDB export, job troubleshooting, and cluster upkeep for trillion‑record workloads.
This article surveys the architecture of cluster resource managers and task schedulers—covering definitions, design principles, and three main categories (centralized, two‑level, and shared‑state) with concrete examples such as Hadoop YARN, Mesos, Spark Drizzle, Borg and Kubernetes—while highlighting their trade‑offs in scalability, fault‑tolerance, and flexibility.
This article explains Hadoop’s core components, the MapReduce programming model, the detailed shuffle and merge processes, and how YARN replaces the classic JobTracker/TaskTracker design to improve scalability and resource utilization in large‑scale data processing clusters.
This article details Lianjia's journey of designing and implementing a low‑latency, stable real‑time computing platform using Spark Streaming on YARN, covering technical selection, architecture components, version compatibility challenges, exactly‑once semantics, graceful shutdown, Kafka tuning, and future enhancements.
The article introduces ArthurCI, a front‑end continuous‑integration platform developed by 58, detailing its design, performance optimizations such as yarn caching and parallel webpack compression, ease‑of‑use integration steps, stability features, and future data‑driven enhancements, while comparing it with tools like TravisCI.
This article explores Hadoop YARN’s evolution, multi‑tenant design, queue and node‑label scheduling, real‑world resource allocation challenges, and data‑driven tools that automate diagnostics and visualizations to optimize CPU and memory usage across massive clusters.
This guide walks through preparing three CentOS 6.9 nodes, configuring hostnames, time sync, password‑less SSH, disabling IPv6, installing JDK, downloading CDH 5.4, setting up core‑site and hdfs‑site XML files, formatting the NameNode, starting HDFS services, configuring YARN and MapReduce, and verifying the installations via the Web UI.
This article shares a comprehensive set of practical tips and solutions for common Spark SQL issues—including out‑of‑memory errors, UDF‑induced GC, thread blocking, system‑property initialization, speculation side‑effects, accumulator traps, concurrent job scheduling, and excessive logging—helping engineers improve stability and efficiency of their Spark‑based financial systems.
This article describes how Ctrip migrated its large‑scale real‑time platform from JStorm to Spark Streaming, detailing the architectural design, the Muise Spark Core encapsulation, operational metrics, encountered pitfalls, and future plans to adopt Flink and Beam for streaming workloads.
This article provides a comprehensive, step-by-step tutorial for setting up a high‑availability Hadoop cluster, covering user creation, JDK installation, host configuration, SSH setup, firewall and SELinux adjustments, Zookeeper deployment, HDFS and YARN HA configuration, essential XML files, and failover testing.
This article explains why package managers are essential, showcases Yarn's step‑by‑step tutorial for creating a simple package manager, and highlights how the resulting tool handles classic challenges like circular dependencies and file‑structure optimization.
npm 5 introduces automatic package‑lock generation, default --save, enhanced Git and file‑dependency handling, new prepack/postpack scripts, stronger integrity checks, a fully managed cache and registry tweaks, while narrowing Yarn’s speed advantage despite early bugs, making it a compelling alternative for npm‑centric workflows.
This article presents eight practical recommendations for Node.js developers, covering dependency locking, lifecycle scripts, modern JavaScript, promises with async/await, code formatting with Prettier, continuous integration testing, security headers via Helmet, and serving over HTTPS.
This article reviews the major Node.js events of 2016—including version updates, the left‑pad controversy, Yarn, Chrome DevTools debugging, and ecosystem tools—describes common application scenarios and framework selection criteria, and offers predictions for Node.js development in 2017.
iQIYI’s Hadoop platform, built since 2010, now spans over a thousand nodes and 60 PB storage, detailing its architectural evolution, operational management practices, encountered challenges, and the custom Gear workflow system that streamlines job scheduling, dependencies, and alerts for large‑scale data processing.
This article provides a comprehensive guide to Inceptor Server HA, covering its high‑availability architecture, configuration steps, required parameters, connection strings, failover behavior, and the HA Tools utility for monitoring and managing master‑standby switches.
YARN’s ResourceManager is a single point of failure, so this guide explains how to configure active/standby mode using Zookeeper, covering leader election, automatic failover, handling false‑dead scenarios, and the essential Zookeeper features such as temporary nodes, watches, and ACLs.
The article details how Facebook built Yarn to overcome npm’s consistency, security, and speed limitations, describing the evolution of their package‑management workflow, Yarn’s lockfile architecture, parallel installation process, additional features, production adoption, and simple commands to get started.
The article examines Yarn, Facebook's new JavaScript package manager, highlighting its focus on speed and version control, its compatibility with npm and Bower, the mixed reactions from developers, and the uncertainty about its long‑term impact on the ecosystem.
This article provides a comprehensive, hands‑on tutorial for setting up a Hadoop 2.6.4 cluster on a CentOS 6.5 development server, covering SSH password‑less login, user/group creation, DNS configuration, JDK installation, environment variables, Hadoop installation, HDFS and YARN configuration, and troubleshooting native library warnings.
This article explains Hadoop 2.0's built‑in MRv2 log collection mechanism, detailing job‑run and task‑run logs, their generation steps, log aggregation, and the role of the JobHistory Server for centralized analysis.
Hadoop 3.0, built on JDK 1.8, adds erasure‑coded HDFS, multi‑NameNode support, native MapReduce task optimizations, cgroup‑based YARN memory and disk isolation, and container resizing, with an alpha slated for summer and a GA release expected in November or December.
The article details the author’s experience upgrading a production Spark cluster from version 1.4.1 to 1.6.1, exposing memory‑spill, unified memory, BlockManager deadlock, Yarn‑kill, UI quirks, and Spark‑SQL compatibility issues, and proposes concrete code‑level fixes for each problem.
This article explains the end‑to‑end workflow of Spark on YARN, covering client initialization, ApplicationMaster actions, driver and executor roles, RDD fundamentals, SparkSQL processing, and practical code examples for building and tuning distributed Spark jobs.
This article introduces Apache Spark’s core architecture, explains how Spark runs on YARN, details driver and executor roles, describes RDD concepts and dependencies, and outlines SparkSQL’s schema‑based query processing, providing code examples for HiveContext and JDBC integration.
This article explains Apache Spark’s core concepts, the RDD programming model, how Spark runs on YARN with driver and executor nodes, the distinction between transformations and actions, partitioning strategies, and an overview of SparkSQL processing.
The article explains how Hadoop clusters suffer from resource contention among multiple users, why YARN alone often fails to prioritize workloads, and how Pepperdata provides deeper visibility and automatic adjustments that reduce low‑priority usage, cut node count, and lower cloud costs.
This article recounts how TalkingData progressively introduced Spark into its Hadoop‑YARN based mobile big‑data platform, detailing early architectures, migration challenges, performance gains, the fully Spark‑centric redesign with Kafka and Spark Streaming, encountered pitfalls, and future plans for further optimization.
Voidbox integrates Docker containers with YARN to simplify distributed application development, improve deployment, boost cluster efficiency, and provide fault‑tolerant, DAG‑based execution modes, enabling seamless resource management for Hadoop‑based big data jobs.
This article walks through the essential steps to install and configure Hadoop 2.0 on a two‑node Linux cluster, covering version selection, directory setup, core XML files, YARN settings, service startup, verification commands, and basic troubleshooting tips.