USDB：ByteDance 针对下一代推荐系统的统一序列存储扩展实践

Authors:Xiaoran Zhu, Jieming Cai, Weiping Yu, Xi Chen, Shu Cai, Jiahao Li, Yifan Xiao, Yan Deng, Yisha Zhou, Yu Lin, Shiru Ren, Siqiang Luo, Peng Xu, Junjun Chai

View PDF HTML (experimental)

Abstract:Recommendation models consume increasingly long, attribute-rich user histories, making small continuous updates and repeated large reads a central storage asymmetry. We present USDB, ByteDance's unified sequence storage practice for streaming and batch workloads. CascadeLSM time-based compaction stages materialize application-specific histories; differentiated replicas share persistent files; partial dumps reconstruct inference inputs from captured recent state and versioned history; and a hybrid columnar format accelerates projected reads. These designs reuse maintenance results across ingestion, inference, and training. Production migration records show 85.8% lower mean write latency, while normalized deployment accounting shows 42.1% lower infrastructure cost. Controlled comparisons with an Optimized RocksDB sequence baseline show faster long-history reads across projections and client loads. Aggregation pushdown sustains 8.5x the throughput of ordinary reads under a constrained receive budget. Production reconstruction and direct-file training results demonstrate how shared sequence maintenance supports fresh serving and repeated historical access.

Comments:	17 pages, 18 figures, 5 tables
Subjects:	Databases (cs.DB)
Cite as:	arXiv:2611.02618 [cs.DB]
	(or arXiv:2611.02618v1 [cs.DB] for this version)
	https://doi.org/10.48550/arXiv.2611.02618 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Weiping Yu Mr. [view email]
[v1] Mon, 2 Nov 2026 06:45:03 UTC (432 KB)