Part 1 of 5 · 1 min read
Why this page comes first
In this module you will
- Explain the wide-column layout, and how it differs from a relational table's fixed columns.
- Explain why column-family stores handle very high write throughput well, connecting back to LSM-trees from Level 5.
- Explain how a compound key lets a column-family store scan a range efficiently within one partition.
- State what a column-family store gives up compared to a relational table.
Level 5 covered LSM-trees and SSTables, a write-optimized storage structure. Column-family stores, such as Cassandra and HBase, are built directly on that same idea. This page covers the data model layered on top of it, and why that combination suits very high write throughput.