一种基于文件系统特征的SLC缓存方案  

AN SLC CACHING SCHEME BASED ON FILE SYSTEM CHARACTERISTICS

在线阅读下载全文

作  者:李明江 陈仁[2] Li Mingjiang;Chen Ren(Institute of Artificial Intelligence and Big Data Application Technology,School of Computer and Information,Qiannan Normal College for Nationalities,Duyun 558000,Guizhou,China;Key Laboratory of Infrared System Detection and Imaging Technology,Shanghai Institute of Technical Physics,Chinese Academy of Sciences,Shanghai,200083,China)

机构地区:[1]黔南民族师范学院计算机与信息学院人工智能与大数据应用技术研究所,贵州都匀558000 [2]中科院上海技术物理研究所中国科学院红外探测与成像技术重点实验室,上海200083

出  处:《计算机应用与软件》2022年第10期224-229,共6页Computer Applications and Software

基  金:上海市科学技术委员会科研计划项目(16XD1404100)。

摘  要:SLC缓存策略广泛应用于基于TLC闪存的固态存储设备中以提升系统的突发写入性能,但它会给系统带来额外的写放大,导致系统寿命减少;当SLC缓存空间耗尽的时候,过早的垃圾回收会导致系统写入性能发生断崖式下降。提出基于文件系统特征的SLC缓存方案,根据文件系统的不同数据类型特点,尤其是利用日志数据的特征,提出独立的日志SLC缓存空间用以写入日志数据,热数据写入到另外的SLC缓存,冷数据则被直接写入到TLC主空间。实验结果表明,该方案能提升系统性能10%~50%,减少写放大10%~30%。SLC caching strategy is widely used in solid-state storage devices based on TLC flash memory to improve the burst write performance of the system,but it will bring additional write amplification to the system and result in the reduction of system life.When the SLC cache space is used up,early garbage collection will lead to a precipitous decline in the system write performance.This paper proposes an SLC caching scheme based on the file system characteristics.According to the characteristics of different data types of the file system,especially the characteristics of the log data,an independent log SLC cache space was proposed to write the log data,the hot data was written to another SLC cache,and the cold data was directly written to the TLC flash space.The experimental results show that the scheme can improve system performance by 10%~50%,and reduce write amplification by 10%~30%.

关 键 词:固态硬盘 SLC缓存 TLC闪存 写放大 性能 

分 类 号:TP311.52[自动化与计算机技术—计算机软件与理论]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象