Intro; Table of Contents; About the Authors; About the Technical Reviewer; Acknowledgments; Foreword; Chapter 1: Introduction to Enterprise Data Lakes; Data explosion: the beginning; Big data ecosystem; Hadoop and MapReduce
Early days; Evolution of Hadoop; History of Data Lake; Data Lake: the concept; Data lake architecture; Why Data Lake?; Data Lake Characteristics; Data lake vs. Data warehouse; How to achieve success with Data Lake?; Data governance and data operations; Data democratization with data lake; Fast Data
Life beyond Big Data; Conclusion.
Chapter 2: Data lake ingestion strategiesWhat is data ingestion?; Understand the data sources; Structured vs. Semi-structured vs. Unstructured data; Data ingestion framework parameters; ETL vs. ELT; Big Data Integration with Data Lake; Hadoop Distributed File System (HDFS); Copy files directly into HDFS; Batched data ingestion; Challenges and design considerations; Design considerations; Commercial ETL tools; Real-time ingestion; CDC design considerations; Example of CDC pipeline: Databus, LinkedIn's open-source solution; Apache Sqoop; Sqoop 1; Sqoop 2; How Sqoop works?
Sqoop design considerationsNative ingestion utilities; Oracle copyToBDA; Greenplum gphdfs utility; Data transfer from Greenplum to using gpfdist; Ingest unstructured data into Hadoop; Apache Flume; Tiered architecture for convergent flow of events; Features and design considerations; Conclusion; Chapter 3: Capture Streaming Data with Change-Data-Capture; Change Data Capture Concepts; Strategies for Data Capture; Retention and Replay; Retention Period; Types of CDC; Incremental; Bulk; Hybrid; CDC
Trade-offs; CDC Tools; Challenges; Downstream Propagation; Use Case.
Centralization of Change DataAnalyzing a Centralized Data Store; Metadata: Data about Data; Structure of Data; Privacy/Sensitivity Information; Special Fields; Data Formats; Delimited Format; Avro File Format; Consumption and Checkpointing; Simple Checkpoint Mechanism; Parallelism; Merging and Consolidation; Design Considerations for Merge and Consolidate; Data Quality; Challenges; Design Aspects; Operational Aspects; Publishing to Kafka; Schema and Data; Sample Schema; Schema Repository; Multiple Topics and Partitioning; Sizing and Scaling; Tools; Conclusion.
Chapter 4: Data Processing Strategies in Data LakesMapReduce Processing Framework; Motivation: Why MapReduce?; MapReduce V1 Refresher and Design Considerations; Yet Another Resource Negotiator
YARN; YARN concepts; Hive; Hive
Quick Refresher; Hive Components; Hive Metastore (a.k.a. HCatalog); Hive
Design Considerations; Hive LLAP; Apache Pig; Pig Execution Architecture; Apache Spark; Why Spark?; Resilient Distributed Datasets (RDD); RDD Runtime Components; RDD Composition; Datasets and DataFrames; Bucketing, Sorting, and Partitioning; Deployment Modes of Spark Application.
