Why Database Storage Engines Need Log-Structured Merge Trees Titelbild

Why Database Storage Engines Need Log-Structured Merge Trees

Why Database Storage Engines Need Log-Structured Merge Trees

Jetzt kostenlos hören, ohne Abo

Details anzeigen
In this episode of Database Tech with Fexingo, Lucas and Luna dive into log-structured merge trees (LSM trees) and why they underpin modern write-heavy databases like Cassandra, LevelDB, and RocksDB. They unpack the trade-offs between LSM trees and traditional B-trees, focusing on write amplification, compaction strategies, and real-world performance implications. Using a concrete example of a time-series metrics system ingesting one million data points per second, they explain how LSM trees turn random inserts into sequential writes and why that matters at scale. Lucas breaks down the compaction process — level-based vs. size-tiered — and Luna questions when LSM trees might actually hurt you. The episode also touches on how compactions cause read stalls and space amplification, and why some databases now mix B-tree and LSM approaches. Perfect for engineers tuning storage backends or choosing a database for high-ingestion workloads. #LSMTrees #DatabaseEngines #StorageEngines #WriteAmplification #Compaction #LevelDB #RocksDB #Cassandra #BTree #TimeSeries #HighThroughput #SequentialWrites #RandomWrites #DatabaseInternals #Technology #DataEngineering #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo
adbl_web_anon_alc_button_suppression_t1
Noch keine Rezensionen vorhanden