Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn about PIM-tree, a novel skew-resistant indexing structure designed specifically for processing-in-memory architectures, in this 37-minute conference talk by Hongbo Kang from Tsinghua University presented at the Simons Institute. Explore how traditional indexing methods face challenges in processing-in-memory environments and discover how PIM-tree addresses data skew issues that can significantly impact performance in memory-centric computing systems. Gain insights into the design principles, implementation strategies, and performance characteristics of this specialized index structure that optimizes data access patterns for processing-in-memory workloads. Understand the theoretical foundations and practical applications of skew-resistant indexing in modern memory management systems, and examine how PIM-tree contributes to more efficient data processing in next-generation computing architectures where computation occurs directly within memory modules.
