Apache Iceberg - Transactional Storage for Large Datasets

Transactional guarantees offered by traditional relational databases has been mostly missing in the world of large data volumes processed by tools emerging from the Hadoop ecosystem, making it harder to construct and warehouse high quality, correct datasets. In recent years, a set of open source table formats emerged to bridge the gap and bring those useful abstractions into the complex reality of distributed data processing. This presentation will focus on one of them, Apache Iceberg, and guide the listeners through the technical details of the format and practical aspects of integrating it with new and existing data engineering workflows.