LeaFTL: A Learning-Based Flash Translation Layer for Solid-State Drives
LeaFTL: A Learning-Based Flash Translation Layer for Solid-State Drives
复制标题
DOI：
10.1145/3575693.3575744
复制
发表时间：
2022-12
期刊：
影响因子：
--
通讯作者：
Jinghan Sun;Shaobo Li-;Yunxin Sun;Chao Sun;D. Vučinić;Jian Huang
中科院分区：
文献类型：
--
作者：
Jinghan Sun;Shaobo Li-;Yunxin Sun;Chao Sun;D. Vučinić;Jian Huang
在现代固态硬盘(SSD)中，闪存页面的索引是其存储控制器中的关键组件。它不仅影响数据访问性能，而且决定了宝贵的设备内DRAM资源的使用效率。已经提出了各种地址映射方案和优化。然而，它们中的大多数都是用人类驱动的启发式方法开发的。本文提出了一种基于学习的闪存转换层LeaFTL，该层在运行时通过线性回归学习地址映射以容忍动态数据访问模式。通过将一大组映射条目分组到一个学习段中，显著减少了地址映射表的内存占用，从而进一步有利于SSD控制器中的数据缓存。LeaFTL还采用了各种优化技术，包括带外元数据验证以容忍错误预测、优化的闪存分配以及学习索引段的动态压缩。我们使用经过验证的固态硬盘模拟器和真正的开放通道固态硬盘板来实施LeaFTL。在不同存储负载下的测试表明，LeaFTL与现有的FTL方案相比，映射表的内存消耗平均降低2.9倍，存储性能平均提高1.4倍。
In modern solid-state drives (SSDs), the indexing of flash pages is a critical component in their storage controllers. It not only affects the data access performance, but also determines the efficiency of the precious in-device DRAM resource. A variety of address mapping schemes and optimizations have been proposed. However, most of them were developed with human-driven heuristics. In this paper, we present a learning-based flash translation layer (FTL), named LeaFTL, which learns the address mapping to tolerate dynamic data access patterns via linear regression at runtime. By grouping a large set of mapping entries into a learned segment, it significantly reduces the memory footprint of the address mapping table, which further benefits the data caching in SSD controllers. LeaFTL also employs various optimization techniques, including out-of-band metadata verification to tolerate mispredictions, optimized flash allocation, and dynamic compaction of learned index segments. We implement LeaFTL with both a validated SSD simulator and a real open-channel SSD board. Our evaluation with various storage workloads demonstrates that LeaFTL saves the memory consumption of the mapping table by 2.9× and improves the storage performance by 1.4× on average, in comparison with state-of-the-art FTL schemes.