Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn about creating lightweight sample tables from larger datasets and their use in statistical inference through this technical presentation from IBM engineers. Explore techniques for generating row samples that help derive table statistics without performing full table scans, ultimately improving Presto's query optimization capabilities. Discover how these sampling methods aid the query planner in making more efficient decisions during the planning phase, leading to better query performance. Gain insights into the implementation details of sampling mechanisms within the Iceberg table format and their integration with Presto's optimization framework.
