A Preliminary Study: Towards Parallel Garbage Collection for NAND Flash-Based SSDs

Guangyu Zhu,Jaehyun Han,Yongseok Son

doi:10.1109/access.2020.3043123

Guangyu Zhu, Jaehyun Han + Show 1 more

Open Access

https://doi.org/10.1109/access.2020.3043123

Copy DOI

Abstract

NAND Flash-based solid-state drives (SSDs) have been widely used as secondary storage devices due to their faster access speed, lower power consumption, and higher reliability compared with hard disk drives. However, application I/O performance can be significantly affected by garbage collection (GC) inside SSDs. For example, a GC operation usually moves valid pages from a victim block into a clean block in a serialized manner. Thus, it increases the GC time and affects the application I/O performance. To address this issue, this article presents a preliminary study on a parallel GC scheme for flash-based SSDs. In our scheme, we parallelize valid page migrations during a GC operation to reduce the total GC time. To do this, we first propose a new flash chip architecture that enables valid page migrations in parallel. Second, we collect information such as new addresses for the migration of valid pages by considering the restriction of SSD operations. Finally, we employ multiple worker threads to migrate the valid pages using the collected information in a parallel manner. We implement and evaluate our scheme using Disksim with Microsoft SSD extension. The experimental result shows that the proposed scheme reduces the overall GC time by 70% and 57% on average compared with the existing scheme and a state-of-the-art GC scheme IPPBE, respectively.

Highlights

T ODAY, NAND flash-based solid state drives (SSDs) have become an important part of storage devices because of their better shock resistance, higher I/O throughput, and lower latency compared with hard disk drives (HDDs) [24], [48]
When using multi-bit per cell flash chips (e.g., TLC and QLC) that have longer read/write time compared with single-bit per cell (SLC) chips, the large number of valid pages further increases the time required for garbage collection (GC) operations
In order to utilize high-performance multi-core processors more efficiently and address the performance degradation introduced by GC, in this article, we propose a page-level parallel garbage collection scheme to reduce the time required for GC

Summary

Introduction

T ODAY, NAND flash-based solid state drives (SSDs) have become an important part of storage devices because of their better shock resistance, higher I/O throughput, and lower latency compared with hard disk drives (HDDs) [24], [48]. Due to an erasebefore-write characteristic of NAND flash memory [4], [47], SSDs perform out-of-place updates. The response time of some I/O requests can reach up to tens of milliseconds [12] when GC takes place in certain address spaces. This is because it blocks incoming I/O requests to those address spaces which increases the I/O response time. HIL connects the host and the SSD together It receives I/O requests from a host, passes them to the SSD, and sends responses to the host.

Methods

Results

Conclusion