Dive into the fascinating world of the "One Billion Row Challenge" that captivated the Java community in this 49-minute Devoxx conference talk. Explore the intricacies of aggregating temperature values from a massive 13 GB CSV file containing one billion rows, grouped by weather station names. Discover the innovative techniques employed by the fastest solutions to process this enormous dataset in under two seconds. Learn about advanced optimization strategies, including parallelization, efficient memory access, optimized parsing routines using SIMD and SWAR, and custom map implementations. Gain insights into the personal experiences and valuable lessons learned by the speakers while organizing and participating in this community-driven challenge. Enhance your understanding of high-performance Java programming and data processing techniques through this engaging presentation.
Overview
Syllabus
1BRC – Nerd Sniping the Java Community by Roy van Rijn, Gunnar Morling
Taught by
Devoxx