How to Write to SSDs

2026-03-10

AI Summary

This work addresses the severe write amplification caused by in-place updates in traditional databases, which significantly limits SSD performance and endurance. It presents the first systematic integration of log-structured (out-of-place) writing into OLTP databases, co-designing the database and SSD layers to reimplement the B-tree-based LeanStore system with native support for emerging SSD interfaces such as ZNS and FDP. Key innovations include a redesigned B-tree structure, optimized write amplification reduction techniques, and seamless adaptation to next-generation storage interfaces. Experimental results demonstrate substantial improvements: on YCSB-A, throughput increases by 1.65–2.24× while flash writes per transaction decrease by 6.2–9.8×; on TPC-C with 15,000 warehouses, throughput improves by 2.45× and flash writes are reduced by 7.2×.

Technology Category

• Search and Optimization: Distributed Search
• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Security and Privacy: Data transparency and provenance
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

This paper demonstrates that adopting out-of-place writes is essential for database systems to fully leverage SSD performance and extend SSD lifespan. We propose a set of out-of-place optimizations that collectively reduce write amplification across both the DBMS and SSD layers. We redesign the in-place, B-tree-based LeanStore to write out-of-place and support these optimizations, and evaluate it on diverse OLTP benchmarks, dataset sizes, and SSDs. The final design improves throughput by 1.65-2.24x and reduces flash writes per transaction by 6.2-9.8x on YCSB-A. On TPC-C with 15,000 warehouses, throughput improves by 2.45x while flash writes decrease by 7.2x. Finally, we show that the architecture can seamlessly support novel SSD interfaces such as ZNS and FDP.

Problem

• SSD
• write amplification
• database systems
• out-of-place writes
• storage optimization

Innovation

• out-of-place writes
• write amplification
• SSD optimization
• LeanStore
• ZNS/FDP