Lecture

LSM Trees: Performance Analysis and Design Choices

Description

This lecture delves into the analysis of the performance of Log-Structured Merge (LSM) trees, focusing on the trade-offs between write throughput and read speed. The instructor discusses the design choices involved in LSM trees, such as workload characterization, multi-level data structures, and the impact of storage heterogeneity. Various strategies to enhance performance, including thread local arrays and skip-list designs, are explored. Experiments with different setups, like Intel Xeon CPUs and varying cache sizes, are presented to showcase the practical implications of these design decisions. The lecture concludes with insights on leveraging Bloom filters and leveled merges to optimize LSM tree performance.

This video is available exclusively on Mediaspace for a restricted audience. Please log in to MediaSpace to access it if you have the necessary permissions.

Watch on Mediaspace
About this result
This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.