WCSE 2017
ISBN: 978-981-11-3671-9 DOI: 10.18178/wcse.2017.06.011

A Progressive Group-Based Wear-Leveling Algorithm for Solid State Drives

Haizhen He, Lei Wen, Rongming Luo

Abstract— Flash-based SSD have the characteristics of non-volatile, low energy consumption, strong shock resistance, etc. However, due to the inherent defects of flash memory, the blocks composed the flash have a limited number of erase cycles. Once reached its erasing limit, the data on the storage blocks will become unreliable. In order to improve the endurance of flash memory, this paper proposed a progressive group-based wear-leveling algorithm which can notably reduce the consumption of RAM space required by wear leveling algorithm while maintaining the wear leveling performance. Our algorithm also provides a new way to identify the cold data more accurately which dramatically minimize the performance overheads due to data migration. The evaluation results reveal that our progressive group-based wear-leveling algorithm consumes only 8% of memory space compared to the previous scheme that manages per-block information, while showing roughly the same wear-leveling performance.

Index Terms— memory consumption, endurance, wear-leveling, performance overheads

Haizhen He, Lei Wen, Rongming Luo
College of Computer Science and Electronic Engineering, Hunan University, CHINA


Cite: Haizhen He, Lei Wen, Rongming Luo, "A Progressive Group-Based Wear-Leveling Algorithm for Solid State Drives," Proceedings of 2017 the 7th International Workshop on Computer Science and Engineering, pp. 65-72, Beijing, 25-27 June, 2017.