The whitepaper “From ClickHouse to ByteHouse” details ByteDance’s large‑scale deployment of ClickHouse, the challenges of moving it to production, and the key optimizations ByteHouse introduces—including custom table engines, a revamped query optimizer, and elastic compute‑storage separation—to achieve petabyte‑level OLAP performance.
Apache Druid is a distributed, column‑store OLAP engine designed for massive real‑time data ingestion and sub‑second queries; this article explains its LSM‑tree‑inspired architecture, DataSource and Segment structures, memory‑based querying, practical deployment steps, common pitfalls, and optimization techniques for high‑throughput analytics.
The article traces the evolution of data platforms from early data warehouses—using schema‑on‑write, columnar storage, and MPP engines—to data lakes that retain raw data with schema‑on‑read, and finally to lakehouse architectures that merge storage and compute, offering unified metadata, versioning, and support for BI, big‑data, AI, and HPC workloads.
The Qunhe Technology data team replaced their legacy Hadoop and Presto clusters with a StarRocks MPP database, achieving up to three times faster queries, supporting billion‑row tables and sub‑second latency for both real‑time and analytical workloads on a daily 1TB data influx.
This article describes how the gaming company Boke City rebuilt its comprehensive data service platform by replacing a CDH‑based Impala solution with StarRocks, detailing the architectural changes, performance benchmark results, and the resulting improvements in query speed, real‑time data updates, and operational simplicity.
This article explains the fundamental differences between transactional (OLTP) and analytical (OLAP) database workloads, describes how data warehouses isolate analytical queries, and introduces star and snowflake schema designs for efficient reporting and business intelligence.
This guide explains ClickHouse fundamentals, OLAP versus OLTP concepts, columnar storage benefits, core performance techniques, the MergeTree family and its indexing, specialized table engines, installation on Linux, Docker deployment, and integration with HDFS, MySQL, and Kafka for modern analytical workloads.
This article explains how to construct a real‑time data warehouse by combining Flink CDC for end‑to‑end change data capture with StarRocks' high‑performance OLAP engine, detailing the architectural challenges, optimization techniques, and a practical e‑commerce case study.
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 explores how combining Flink CDC with StarRocks can streamline real‑time data pipelines, reduce component complexity, support both full and incremental synchronization, and enable efficient OLAP queries and updates for fast, scalable analytics across diverse business scenarios.
Apache Kylin leverages pre‑computed OLAP cubes on Hadoop/Spark/Flink to deliver sub‑second query responses for massive datasets, detailing its architecture, integration with BI platforms, user security, cube building, monitoring, and storage using HBase, illustrating how it overcomes big‑data analytical challenges.
The article explains the challenges of big‑data analytics, introduces pre‑computation OLAP concepts, and details how Apache Kylin together with Spark, Flink, Presto and other components can be integrated into a BI platform to achieve near‑real‑time query performance on massive datasets.
During the 2022 China Internet Industry Application Salon, JD Cloud’s product manager explained how the Color gateway, an API gateway handling billions of daily requests, overcomes stability, high‑availability, reliability, and performance challenges during peak sales by adopting a cloud‑native ClickHouse data warehouse that boosts processing speed, reduces costs, and provides real‑time analytics.
This article explains the design and implementation of StarRocks 2.x Primary Key tables, covering real‑time update mechanisms, write and commit workflows, in‑memory primary indexing, compaction, read‑path optimizations, performance benchmarks, and upcoming features such as partial and conditional updates.
Facing massive data storage and query challenges, 360 upgraded its analytics architecture by adopting StarRocks, achieving multi‑dimensional, high‑concurrency analysis, simplified data pipelines, and significant performance and cost improvements across its radar and user‑portrait platforms.
The article details ByteDance's large‑scale deployment of the Presto OLAP engine for ad‑hoc, BI, and near‑real‑time analytics, describing its architecture, multi‑coordinator high‑availability design, routing gateway, adaptive cancel, history server, materialized‑view support, Hudi connector integration, and how these innovations improve performance, stability, and operational efficiency.
This article provides a comprehensive overview of ClickHouse, covering its core use cases, detailed explanations of the various table engines, recommended schema and deployment patterns, performance‑tuning parameters, tooling choices, and operational guidelines for building and maintaining high‑availability OLAP clusters.
This article provides a comprehensive overview of ClickHouse, an open‑source column‑oriented MPP analytical database, covering its advantages and drawbacks, key features, typical use cases, data access flow, installation steps, core directories, indexes, data types, database and table engines, as well as detailed cluster architecture and deployment patterns.
This article describes how ZhongAn Insurance’s JiZhi data‑analysis platform migrated from ClickHouse to the MPP OLAP engine StarRocks, detailing the business requirements, architectural challenges, benchmark results across single‑table and multi‑table queries, and the resulting improvements in latency, concurrency, and operational simplicity for real‑time analytics.
This article presents a technical case study of ZhongAn's Jizhi analytics platform, detailing how switching from ClickHouse to the MPP OLAP engine StarRocks resolved multi‑concurrency and join performance bottlenecks, improved real‑time query speed, and enabled near‑billion‑row data handling for insurance business operations.
After months of using MongoDB for frontend monitoring logs, the data grew to billions of records causing severe query slowdown, prompting a migration to ClickHouse where columnar storage, partitioning, and OLAP capabilities dramatically improved performance and storage efficiency.
This article explains OLAP as a core data‑warehouse component that enables fast, multidimensional analysis for business intelligence, describes the structure of OLAP cubes, outlines basic operations such as drill‑down, roll‑up, slice‑and‑dice and pivot, compares MOLAP, ROLAP and HOLAP, and discusses OLAP’s relationship with OLTP and cloud architectures.
Facing inflexible point‑lookup limits and slow query times on HBase, Zepp Health redesigned its massive event‑tracking data pipeline—migrating ingestion through Kafka, Flink, and Hudi to a StarRocks‑based OLAP layer—achieving sub‑100 ms average query latency, 20 % storage savings, and dramatically faster multi‑dimensional analytics.
This article recounts the author's journey from using MongoDB for front‑end monitoring logs to migrating to ClickHouse, detailing the challenges with large‑scale data, optimization attempts, the fundamental differences between row‑ and column‑oriented databases, and the resulting performance and storage improvements.
This talk explains why observability is essential for cloud‑native IT systems, covering its core value of empowerment, various definitions, evaluation criteria such as zero‑intrusion, multidimensionality and real‑time response, and practical building approaches using SaaS, open‑source and integration, illustrated with numerous industry case studies.
This guide provides a comprehensive overview of ClickHouse, covering its core use cases, detailed table‑engine choices, cluster design, Zookeeper integration, query and data‑loading best practices, client tools, and key configuration parameters to ensure high performance and reliability in OLAP workloads.
This article presents a comprehensive design of a real-time advertising data warehouse powered by Flink, covering construction background, technical and data‑warehouse architecture, real‑time OLAP, stability and data‑quality guarantees, future plans, and the integration of Hologres for simplified processing.
The article examines the data‑storage problems caused by rapid microservice growth, explains why traditional Hadoop/Spark stacks were rejected, presents benchmark comparisons that show ClickHouse’s superior performance and compression, and details practical ClickHouse deployment, schema design, sharding, TTL, indexing, and monitoring integrations for real‑time analytics.
Shopee’s engineering team scaled a 100‑node Apache Druid cluster for real‑time OLAP by redesigning the Coordinator load‑balancing algorithm, adding incremental metadata pulls, introducing a segment‑merged result cache, and building exact‑count and flexible sliding‑window operators, while planning cloud‑native deployment.
This article explains the evolution, challenges, and technical solutions for building an end‑to‑end real‑time data warehouse by combining Apache Flink's stream processing with StarRocks' ultra‑fast OLAP engine, covering architecture, data models, integration methods, best‑practice cases, and future roadmap.
The article describes how Haodf's SRE team replaced Elasticsearch with ClickHouse to handle massive microservice logs, achieve low‑latency queries, reduce storage costs, and support real‑time monitoring, tracing, and metric analysis through columnar OLAP features, sharding, TTL, and materialized views.
ByteDance’s internal Presto platform, supporting nearly one million daily queries across ad‑hoc, BI visualization, and near‑real‑time analytics, achieves high performance and stability through SparkSQL compatibility, multi‑Coordinator architecture, dynamic routing, adaptive query cancellation, History Server, materialized views, and a dedicated Hudi connector.
This article analyzes the challenges of massive data query efficiency, explains how Flink's stream processing and ClickHouse's OLAP engine complement each other, and presents a layered real‑time data‑warehouse architecture with practical guidance on data ingestion, write strategies, quality assurance, and evolving batch‑stream integration patterns.
The Fast OLAP Forum held on December 19 at DataFunCon gathers leading experts from Baidu, Tencent, JD, and FreeWheel to share cutting‑edge techniques in vectorized execution, cloud‑native ClickHouse, large‑scale OLAP architectures, and Presto optimizations, offering deep insights for practitioners dealing with massive real‑time data workloads.
Autohome replaced Flink and Kylin with StarRocks to power sub‑second real‑time OLAP analytics, detailing data sources, pain points, benchmark comparisons against Apache Kylin, ClickHouse, Presto, Spark, and Doris, integration with Flink‑connector, broker‑load scripts, monitoring setup, and lessons learned from large‑scale deployments.
This article describes AutoHome's real‑time data analysis architecture, the challenges of existing OLAP solutions, the reasons for choosing StarRocks, detailed performance comparisons with Kylin, ClickHouse, Doris, Presto and Spark, and the practical integration of StarRocks with Flink, broker‑load scripts, and monitoring tools.
At Baidu Aifanfan, the data team built a unified real‑time and offline big‑data platform—leveraging Watt, Bigpipe, Fengge, AFS and Palo within Lambda/Kappa patterns and a fast‑slow parallel rollout—that cut OLAP query latency from 18 minutes to under 15 seconds, enabled self‑service analytics, and standardized metrics across 15 agile teams.
SF Express’s data engineering team details how they migrated from a mixed‑component OLAP stack to a unified StarRocks platform, describing the evaluation criteria, performance‑critical design choices, import and query optimizations, and future roadmap for a high‑availability, low‑cost big‑data analytics solution.
This article presents Tencent's game big‑data analysis platform, detailing its background, the architecture of the iData engine—including offline multi‑dimensional analysis (TGMars), online portrait analysis (TGFace), and real‑time multi‑dimensional analysis (TGDruid)—application scenarios, performance insights, and future ecosystem and open‑source plans.
This article compares Apache Doris and ClickHouse across architecture, deployment, multi‑tenant management, data import, storage, query capabilities, performance testing, and cost, providing practical guidance for selecting the most suitable analytical database in large‑scale OLAP scenarios.
This article details Youzan's adoption of ClickHouse for real-time analytics, covering its evolution from Presto, Druid, and Kylin, the system's architecture, deployment strategies, use cases, performance characteristics, limitations, and future roadmap, including integration with Apache Doris and emerging big‑data trends.
This article explains how JD.com designs and evolves its OLAP platform, covering data sources, ingestion, storage, real‑time and offline processing, key challenges such as timeliness, high throughput, consistency, and the solutions implemented to support massive e‑commerce analytics.
This article details Vipshop's OLAP evolution, covering the deployment, optimization, and containerization of Presto, ClickHouse, and Kylin, the challenges faced, self‑developed tooling, and future directions for intelligent scaling and resource management.
JD.com’s customer service team leverages the open‑source MPP database Doris to power real‑time and offline OLAP dashboards, detailing data ingestion pipelines, full‑link monitoring, dual‑stream high‑availability design, dynamic partition management, multi‑layer caching strategies, and performance optimizations applied during the 2020 11.11 shopping festival.
This article describes how Ctrip's hotel data platform migrated from ClickHouse to StarRocks, detailing the platform's current state, pain points, the evaluation and selection of StarRocks, its architecture, performance benchmarks, data ingestion models, high‑availability design, and future migration plans.
JD's OLAP platform runs on ClickHouse and Doris across 3,000 servers, handling billions of daily queries and petabytes of data, and this article details the selection criteria, cluster deployment models, high‑availability architecture, operational challenges, and future roadmap.
This article explains how Databend, a cloud‑native OLAP data warehouse, addresses modern data‑warehouse challenges by separating storage and compute, providing elastic scaling, multi‑cloud support, and efficient query planning and execution to deliver low‑cost, on‑demand analytics.
This article shares practical experiences with Doris, covering materialized view concepts, typical use cases, index principles, performance optimizations, and real‑world scenarios such as order analysis, PV/UV aggregation, and detailed queries, while also providing operational tips and Q&A insights.
This article presents a detailed technical comparison between Apache Doris and ClickHouse, covering their architecture, deployment, distributed capabilities, transaction support, data import, storage design, query performance, cost, and future development, and provides guidance on selecting the appropriate engine for specific OLAP scenarios.
This article presents an in‑depth overview of ClickHouse Projection, covering its background, core features, architectural details, practical use cases, performance gains, advantages and drawbacks, and real‑world production results, illustrating how Projection enhances OLAP workloads at Kuaishou.
This article shares practical experiences with Doris, covering materialized view concepts, typical use cases, advantages, creation syntax, prefix index principles, performance‑boosting scenarios such as order analysis, PV/UV counting, detail queries, and operational tips for high‑throughput and low‑latency workloads.
This article presents the architecture, key features, and operational benefits of a cloud‑native ClickHouse OLAP platform, describing how storage‑compute separation, a unified master node, and shared storage reduce cost, improve availability, and simplify management while remaining fully compatible with the open‑source ClickHouse ecosystem.
This article presents a comprehensive design of a cloud‑native ClickHouse OLAP system, detailing its three‑layer architecture, storage‑compute separation, unified metadata management, high‑availability mechanisms, elastic scaling, cost reductions, and future enhancements for multi‑replica and MPP query support.
The article details JD.com’s large‑scale OLAP strategy using ClickHouse as the primary engine and Doris as a secondary engine, covering application scenarios, component selection criteria, cluster deployment models, high‑availability architecture, fault‑handling procedures, performance tuning, and future cloud‑native plans.
The article details how Kuayue Group's big‑data center replaced Presto and ClickHouse with DorisDB, achieving sub‑second query latency, simplifying architecture, and enabling both online and real‑time OLAP analytics across millions of daily requests.
This article presents an in‑depth overview of ClickHouse Projection, covering its background, definition, practical use cases, underlying architecture, query analysis, consistency guarantees, performance comparisons, and real‑world production results, highlighting how it enhances OLAP workloads while maintaining strong data consistency.
This article presents an in‑depth overview of ClickHouse's Projection feature, covering its background, definition, storage and query mechanisms, practical use‑case demonstrations, performance comparisons with competing OLAP systems, and real‑world production results that highlight its advantages and limitations.
The article chronicles Lenovo Liancheng Zhida’s three‑stage evolution of OLAP engines—from early SQL Server scripts, through a Hadoop‑based Presto solution, to the adoption of DorisDB—detailing architecture, tool comparisons, implementation practices, and the performance and operational benefits achieved.
The article describes how Beike's data platform consolidated multiple OLAP engines into a single DorisDB‑based solution, achieving higher query performance, lower operational complexity, and robust support for diverse analytics workloads such as metrics, dashboards, real‑time updates, and multi‑table joins.
The article describes how Beike Real Estate consolidated multiple OLAP engines into a single DorisDB‑based platform, detailing the business challenges, DorisDB’s technical advantages, extensive performance and concurrency benchmarks, and the resulting improvements in stability, query speed, and operational simplicity across various business scenarios.
This article explains how data-driven decision making evolves from simple Excel and Python analysis to robust data warehouses, introduces star‑schema dimension modeling, and demonstrates practical integration of Alibaba Cloud MaxCompute with Tableau for interactive OLAP‑style analytics.
This article explains how Beike adopted the Druid OLAP engine to handle massive real‑time and offline query workloads, detailing its four‑component architecture, key technologies such as deep storage and metadata storage, practical optimizations for data ingestion, query caching, dynamic throttling, timeout control, and a roadmap for future enhancements.
This article explains the growing analytical demands in the big‑data era, the limitations of traditional OLAP, and how Kyligence’s distributed OLAP engine addresses data‑island issues, multi‑dimensional and many‑to‑many analysis, unified security, and performance optimization with MDX on Spark, delivering a seamless Excel‑like experience.
This article provides an in‑depth overview of Apache Kylin, covering its history, mission, core MOLAP principles, technical architecture, step‑by‑step installation (Docker and Hadoop), performance tuning, advanced cube settings, and detailed case studies from major companies such as Baidu, Lianjia, and Didi.
This article details Xiaohongshu’s multi‑stage evolution of its OLAP infrastructure—from Redshift to Presto, ClickHouse, and finally DorisDB—describing the data pipeline, tool comparisons, advertising use‑case implementation, and the resulting performance and operational benefits.
This article details JD's OLAP implementation using ClickHouse as the primary engine and Doris as a secondary engine, covering business scenarios, selection criteria, multi‑tenant deployment, high‑availability architecture, encountered challenges, and future roadmap for cloud‑native, scalable analytics.
This article explains the various classifications of OLAP data warehouses—including MOLAP, ROLAP, HOLAP, and HTAP—based on data volume and modeling, reviews common open‑source ROLAP products, and details performance‑boosting techniques such as MPP architecture, cost‑based optimization, vectorized execution, and storage optimizations.
This article details Vipshop's OLAP evolution, describing how Presto, Kylin, and ClickHouse are integrated, the deployment architecture with HAproxy and chproxy, containerization on Kubernetes, and the Flink‑ClickHouse pipeline that enables self‑service analysis of hundred‑billion‑row datasets while addressing performance challenges and future roadmap.
A unified semantic layer for data development solves metric‑change ripple effects, developer burden, and large‑scale query performance problems by offering consistent metric definitions, multi‑view access, concise auto‑generated SQL, instant propagation of updates, and engine‑driven optimal query selection, thereby bridging business and engineering and cutting maintenance effort.
ClickHouse, born from Yandex’s Metrica and now a top‑50 open‑source analytics engine, achieves exceptional speed through a vectorized compute engine, column‑store architecture, and an active community, powering real‑time workloads at companies like Tencent Music, Sina, Bilibili, and Suning while introducing features such as column merging, projections, and storage‑compute separation for future scalability.
The article details how Alibaba’s Data Bank leverages AnalyticDB’s cold‑hot tiered storage, high‑throughput real‑time writes, and low‑latency OLAP capabilities to handle petabyte‑scale consumer data, support flexible AIPL analysis, crowd profiling, and rapid audience selection while cutting costs and ensuring elasticity during peak events.
The article describes how Yuanfudao's data middle platform built a high‑performance OLAP service using the MPP HOLAP engine DorisDB to unify real‑time and batch analytics, meet low‑latency and high‑concurrency requirements, and support diverse education‑industry use cases such as live‑stream monitoring, advertising, and order analytics.
Yuanfudao's data middle platform leverages the MPP database DorisDB to create a unified OLAP system that supports both real‑time and batch analytics, handling millions of queries daily with sub‑second latency while meeting complex business requirements across its education services.
A unified semantic layer for data development creates a consistent, multi‑view representation of metrics that buffers logical changes, lets downstream applications use metric names only, and enables analysts and developers to select optimal query objects, thereby reducing misunderstandings, cutting rework, and improving query performance and maintainability.
This article explains how JD constructs its OLAP platform from data ingestion to storage, querying, and management, describing the diverse data sources, real‑time and offline processing, scalability, consistency, fault tolerance, and future optimization plans, while addressing key technical challenges and solutions.
This article describes how JD.com leverages the open‑source MPP analytical database Doris for real‑time and offline OLAP on customer‑service data, covering data ingestion pipelines, dual‑stream high‑availability design, dynamic partition management, multi‑level caching, monitoring with Prometheus‑Grafana, and performance optimizations applied during major sales events.
This article reviews the evolution of Beike's OLAP platform—from the early Hive‑to‑MySQL stage, through a Kylin‑based architecture, to a flexible multi‑engine solution—detailing the design choices, metric system, engine selection criteria, encountered challenges, and future development plans.
The article details Beike's OLAP platform built on Druid, covering its four‑layer architecture, selection criteria, performance comparisons with Kylin, data ingestion workflows, custom improvements for data import and real‑time distinct counting, and operational measures such as caching, dynamic throttling, and HDFS storage optimization.
This article details Beike's large‑scale OLAP platform, explaining why Druid was chosen over Kylin, describing the platform's four‑layer architecture, presenting performance and storage benchmarks, and outlining practical improvements to data ingestion, real‑time distinct counting, and cluster stability for high‑concurrency business scenarios.
Suning’s data middle platform integrates an accelerated OLAP engine, a star‑schema metrics system, a standardized visualization tool, and a unified report portal to break data silos, enhance security, and transform traditional enterprises into technology‑driven businesses.
This article explains the characteristics of OLTP and OLAP workloads, compares their I/O patterns, and provides Oracle database performance best‑practice guidelines, including storage planning, SAN architecture, operating‑system queue‑depth settings, and SwingBench testing results for optimal configuration.
The article details the design and engineering of a high‑performance path analysis model for internet data operations, explaining session handling, Sankey visualizations, adjacency‑table storage, multi‑granular session partitioning, Spark‑to‑ClickHouse pipelines, and optimizations that enable billion‑scale user‑path queries in about one second.
Suning’s Data Middle Platform integrates an accelerated OLAP engine, a star‑schema metric system, a visualization tool built on standardized dimensions, and a unified report portal to solve data silos, improve security, and enable enterprises to evolve into technology‑driven organizations.
This article details JD Search's journey from a Storm‑based pipeline to a Flink‑driven architecture backed by Apache Doris, covering business requirements, technical challenges, design trade‑offs, performance optimizations for massive traffic spikes, and future plans for their real‑time OLAP data warehouse.
This article explains how Suning's big‑data team incorporated ClickHouse into their end‑to‑end monitoring ecosystem, detailing the architecture, trace‑ID propagation, slow‑query tracking, MergeTree health checks, replica delay analysis, and the role of Chproxy in delivering comprehensive observability for high‑performance OLAP workloads.
This article describes how JD's Energy Management Platform leverages ClickHouse as a high‑performance, MPP‑based OLAP engine to provide real‑time, multi‑dimensional analytics on IoT energy data, covering business background, technology selection, system architecture, data ingestion, storage, replication, and a generic query interface with code examples.
This article describes Beike's adoption of the Druid OLAP engine, detailing its technical architecture, key data structures, offline and real‑time optimization techniques, cluster load‑protection strategies, and the platformization roadmap for scalable analytics across multiple business lines.
This article explains how JD’s Energy Management Platform uses ClickHouse as an MPP‑based OLAP engine to ingest, store, and provide multi‑dimensional real‑time analytics on energy consumption data, covering architecture decisions, data pipelines, replication, sharding, and a generic query interface.
The article explains ClickHouse’s high‑performance columnar OLAP design, its vectorized execution, sparse primary‑key indexes and MergeTree engines, contrasts it with ROLAP/MOLAP approaches, and details Youzan’s large‑scale deployment—including dual‑replica clusters, ingestion pipelines, routing architecture, current challenges, and future container‑based expansion plans.
This article explains the background, challenges, and architectural design of a big‑data OLAP platform that integrates Apache Kylin with a BI system, detailing pre‑computation strategies, cube construction, user authentication, storage engines, and query mechanisms to achieve sub‑second analytics on massive datasets.
ClickHouse is an open‑source column‑oriented OLAP database developed by Yandex that offers real‑time queries, high performance compared to many traditional databases, a rich set of features, and a modular architecture including columns, data types, blocks, parsers, functions, and various table engines.
This article presents a detailed case study of how Meituan's in‑store dining sales team identified severe efficiency issues in their Apache Kylin‑based OLAP system, dissected the construction process, and applied a step‑by‑step optimization roadmap—including engine migration, dimension pruning, resource configuration, and Spark‑based layered building—to boost query performance and achieve near‑perfect SLA.
This article provides a comprehensive introduction to ClickHouse, an open‑source column‑oriented OLAP database, covering its high‑performance benchmarks, core architectural components, query processing model, deployment patterns, Java client usage, and real‑world implementations at large enterprises.
This article introduces ClickHouse as a column‑oriented OLAP database, explains its architectural differences from row‑based systems, details a real‑world migration from MySQL with code examples, showcases performance gains, and shares practical tips and pitfalls for production use.
This article recounts the author’s journey building a real‑time data warehouse using Flink, Kafka, Redis, and ClickHouse, describing the initial batch‑oriented setup, successive architectural evolutions, challenges with wide tables and dimension data, and the final OLAP‑centric solution with secondary caching.