Organizations striving to build applications for streaming data have a new possibility to ponder: the use of ingestion engines at the front end of their Hadoop systems. With this O’Reilly report, you’ll learn how these fast data front ends process data before it reaches the Hadoop Data File System (HDFS), and provide intelligence and context in real time. This helps you reduce response times from hours to minutes, or even minutes to seconds. Author and independent consultant Akmal Chaudhri looks at several popular ingestion engines, including Apache Spark, Apache Storm, and the VoltDB in-memory database. Among them, VoltDB stands out by providing full Atomicity, Consistency, Isolation, and Durability (ACID) support. VoltDB also lets you build a fast data front-end that uses the familiar SQL language and standards. Learn the advantages of ingestion engines as well as the theoretical and practical problems that can come up in an implementation. You’ll discover how this option can handle streaming data, provide state, ensure durability, and support transactions and real-time decisions. Akmal B. Chaudhri is an Independent Consultant, specializing in big data, NoSQL, and NewSQL database technologies. He has previously held roles as a developer, consultant, product strategist, and technical trainer with several blue-chip companies and big data startups. Akmal regularly presents at international conferences and serves on program committees for several major conferences and workshops.
