ByteDance Unveils USDB to Optimize Recommender Storage

1 reports archived1 independent sourcesupdated 11/2/2026, 14:45:03

Synthesis & Latest Updates

1 Sources Cross-Validated

ByteDance introduces USDB, a unified sequence storage system addressing the 'small updates, large reads' asymmetry in recommender models. By leveraging cascade LSM compaction, hybrid columnar formats, and aggregation pushdown, USDB reduces infrastructure costs by 42.1% and mean write latency by 85.8%, while significantly boosting throughput for long-history reads.

LATEST/ByteDance introduces USDB, a unified sequence storage system addressing the 'small updates, large reads' asymmetry in recommender models. By leveraging cascade LSM compaction, hybrid columnar formats, and aggregation pushdown, USDB reduces infrastructure costs by 42.1% and mean write latency by 85.8%, while significantly boosting throughput for long-history reads.

Hourly heat curve

2 fully observed hours

Fewer than 3 fully observed hours; trend not plotted yet.

Coverage timeline

Total 1 reports · Latest first

Hacker News AIT2Nov 2, 14:45·68 pts
ByteDance Unveils USDB to Optimize Recommender Storage
- Addresses storage bottlenecks caused by 'small continuous updates, repeated large reads' of user history sequences
- Utilizes cascade LSM time-based compaction and hybrid columnar formats to accelerate projected reads and inference input reconstruction
- Production data shows an 85.8% reduction in mean write latency and a 42.1% decrease in infrastructure costs
View Related Reports & Analysis Original Source