Iceberg tables running slow? This comprehensive session reveals 8 proven optimization techniques with real benchmark data from production AWS workloads. Discover
how to achieve dramatic performance improvements through: (1) Distribution modes for optimal file layout, (2) Statistics optimization for query pruning, (3) Z-
ordering for multi-dimensional sorting, (4) Compaction strategies to eliminate small files, (5) File size optimization for metadata efficiency, (6) Partition
evolution without data rewrites, (7) Spark Adaptive Query Execution integration, and (8) Puffin files for advanced statistics. Real-world example: 10 billion row,
5TB table optimized from 312 seconds to 18 seconds—a 17x speedup with 93% cost reduction. Each technique includes specific benchmarks, and practical guidance on when to apply them. Learn the optimization decision tree and avoid common pitfalls.