- ホーム
- > 洋書
- > 英文書
- > Computer / General
Full Description
Build AI-native lakehouses with Apache Hudi. Master incremental processing, indexing, and table services, and use AI-assisted tools to design, build, and manage Hudi tables at enterprise scale.
Key Features
Architect production lakehouses with Apache Hudi, from ingestion to reliable operations at petabyte scale
Bring unstructured data into the lakehouse to serve AI alongside analytics and streaming
Apply AI-assisted tools to design pipelines, diagnose issues, and manage table health
Book DescriptionModern lakehouses must go beyond analytics to support continuous updates, diverse workloads, and growing data volumes. With AI introducing new demands for unstructured data management and intelligent operations, building reliable, scalable platforms is more critical than ever.
Apache Hudi provides the transactional foundation for modern lakehouses. Drawing on extensive experience building petabyte-scale data systems and contributing to the Apache Hudi community, the authors share practical insights into designing and operating lakehouses for analytics, streaming, and AI.
You'll learn to build batch, streaming, and CDC pipelines, optimize storage and indexing, and query Hudi tables across engines. Explore compaction, clustering, cleaning, archival, concurrency, recovery, monitoring, and performance optimization.
Through hands-on examples and real-world case studies, discover how to integrate unstructured data, support AI applications, and use AI-assisted tools to design pipelines, assess table health, and troubleshoot issues. Build production-ready lakehouses prepared for evolving data and AI workloads.What you will learn
Design scalable lakehouse architectures with Apache Hudi
Build tables with different layouts like Copy-on-Write and Merge-on-Read
Ingest batch, streaming, and CDC pipelines with Hudi writers
Query Hudi tables across engines and table formats
Optimize storage with compaction, clustering, and cleaning
Accelerate reads and writes using metadata and indexing
Support AI workloads with unstructured data and vector search
Operate and manage Hudi tables with AI-assited tools
Who this book is forThis book is for data engineers, analytics engineers, data platform architects, and technical leads building lakehouse platforms, CDC or streaming pipelines and AI applications. It also serves platform engineers and engineering managers looking to modernize data infrastructure, manage unstructured data, or improve lakehouse operations with AI-assisted tools. Readers will benefit most if they understand data lake concepts and have working familiarity with Apache Spark (Scala or PySpark) and common ETL/ELT workflows.
Contents
Table of Contents
From Data Lakes to Open Lakehouse Platforms
Apache Hudi Essentials and QuickStart
Ingesting Data into the Lakehouse
Querying Hudi Tables Across Engines
Table Services and Storage Optimization
Case Study I: Building a Real-Time Marketplace Payout Lakehouse
Case Study II: Designing a Lakehouse for CDC Workloads
Advanced Capabilities: Indexing, Concurrency, Time Travel
Under the Hood: Hudi Architecture as a Platform Layer
Running, Deploying and Operationalizing Hudi Jobs
Operating Lakehouses with AI-Assisted Skills and Agents
Towards Self-Driving Lakehouses



