Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore Apache Hudi's Non-Blocking Concurrency Control (NBCC) in this 34-minute conference talk that addresses critical challenges in high-throughput streaming data ingestion. Learn how traditional concurrency control mechanisms fail under heavy concurrent writes, causing errors, resource waste, and data staleness, while discovering NBCC as a revolutionary solution for conflict-free, high-throughput ingestion into single tables. Understand the innovative architecture that leverages specialized file layouts, TrueTime semantics, and bucket indexing to ensure efficient event-time ordering and seamless Apache Flink pipeline integration. Master the NBCC architecture and workflow, including its alignment with Flink for streaming ingestion, and discover how file layouts, LSM-tree backed logs, and TrueTime semantics eliminate bottlenecks in concurrent data processing. Follow along with a practical Flink SQL demonstration showcasing concurrent writes across multiple streams, and gain insights into future NBCC enhancements for metadata tables, clustering, and advanced indexing capabilities. Acquire actionable knowledge for building efficient, scalable, and real-time data ingestion systems, whether developing high-throughput streaming pipelines or implementing modern lakehouse solutions that require simultaneous updates, real-time dataset joins, and multiple writers without performance bottlenecks.
