HAML-SSD: A Hardware Accelerated Hotness-Aware Machine Learning based SSD Management
HAML-SSD: A Hardware Accelerated Hotness-Aware Machine Learning based SSD Management
复制标题
DOI：
10.1109/iccad45719.2019.8942140
复制
发表时间：
2019-11
期刊：
影响因子：
--
通讯作者：
Bingzhe Li;Chunhua Deng;Jinfeng Yang;D. Lilja;Bo Yuan;D. Du
中科院分区：
文献类型：
--
作者：
Bingzhe Li;Chunhua Deng;Jinfeng Yang;D. Lilja;Bo Yuan;D. Du
近年来，固态硬盘（SSD）作为一种快速存储设备，在从移动计算到大型分布式系统的许多应用中发挥着重要作用。然而，由于基于 NAND 的闪存的固有特性（包括有限的擦除周期和不对称的写入和擦除操作），SSD 的性能可能会大幅下降。之前的工作将热/冷数据分离到不同的块中，以提高 SSD 性能。 “热度”通常定义为页面的累积更新频率。然而，我们认为，与更新频率相关的“热度”定义中还应该考虑一个额外的新参数，即平均更新时间间隔。此外，为了自适应地对热/冷数据进行分类，应用了机器学习算法来更好地适应动态变化的跟踪 I/O 访问模式。本文提出了一种基于机器学习（ML）的 SSD 管理，称为 HAML-SSD。应用ML算法的目的是基于“热度”的新定义，对具有相似“热度”的数据进行动态聚类。因此，使用二维聚类算法将分类到同一簇中的页面存储在同一块内。此外，为了获得合理的训练时间，SSD中设计了一个称为HAML单元的特定硬件组件。最后，实验结果表明，与之前评估真实痕迹的工作相比，HAML-SSD 的响应时间减少了约 26.3% – 57.7%。
Solid state drive (SSD) as a fast storage device has been playing an important role across many applications from mobile computing to large distributed systems in recent years. However, the performance of the SSD can be degraded tremendously due to the intrinsic properties of NAND-based flash memory including limited erase cycles and asymmetric write and erase operations. Previous works separated hot/cold data into different blocks in order to improve SSD performance. “Hotness” is typically defined as the cumulative update frequencies of pages. However, we believe that an additional new parameter, average update time interval, should also be considered into the “hotness” definition associated with the update frequency. Moreover, to adaptively classify hot/cold data, a machine learning algorithm is applied to better accommodate the dynamically changed I/O access patterns of traces. In this paper, a machine learning (ML) based SSD management called HAML-SSD is proposed. The purpose of applying the ML algorithm is to dynamically cluster the data with similar “hotness” based on a new definition of “hotness”. Thus, a two-dimension clustering algorithm is used for storing the pages categorized into the same cluster within the same block. Moreover, to obtain reasonable training time, a specific hardware component called HAML-unit is designed in the SSD. Finally, the experimental results indicate that the HAML-SSD decreases the response time around 26.3% – 57.7% compared to previous works with the evaluation of real traces.