面向深空探测的高检索精度固态存储系统设计  

An SSD Design with High Index Accuracy for Deep-Space Exploration

在线阅读下载全文

作  者:罗唤霖 叶恒 白郁 杨鹿 孙晶晶 朱新忠 LUO Huanlin;YE Heng;BAI Yu;YANG Lu;SUN Jingjing;ZHU Xinzhong(Shanghai Aerospace Electronic Technology Institute,Shanghai 201109,China;Shanghai Academy of Spaceflight Technology,Shanghai 201109,China)

机构地区:[1]上海航天电子技术研究所,上海201109 [2]上海航天技术研究院,上海201109

出  处:《上海航天(中英文)》2022年第S01期192-197,共6页Aerospace Shanghai(Chinese&English)

摘  要:在深空探测中,探测器上的固态存储系统需要检索出记录的数据并将其传输到地球上。为了更好地利用宝贵的传输时间,研究人员希望该系统具有更准确的检索能力。在现有的探测器固态存储系统设计中,检索精度受到系统RAM容量的限制。提出了一种日志型固态存储系统设计方法,该方法利用日志结构机制对固态存储介质NAND Flash进行管理,以节省RAM容量。通过这种设计方法,可以实现非常小的数据存储粒度,并给每个数据颗粒配备可扩展的索引信息,使得在系统RAM容量有限的情况下可极大提高数据的索引精度。实验结果表明:使用高可靠的反熔丝FPGA实现1 TB固态存储系统,并要求系统可检索到页时,索引数据只需要5.9 Mb的RAM容量开销,系统的平均写速率为118 Mbit/s,平均索引延迟为40.1μs。In deep space exploration,the solid state disk(SSD)system on the probe is required to retrieve the recorded data and transmit it to the Earth.In order to make better use of the precious transmission time,the researchers expect that the system has more accurate retrieval capabilities.In the existing detector SSD system design,the retrieval accuracy is limited by the random access memory(RAM)capacity of the system.In this paper,a log-structure SSD system design method is proposed.The log-structure mechanism is used to manage the NAND Flash of the solid state medium so as to save the RAM capacity.With this design method,very small data storage granularity can be achieved.Each data particle is equipped with expandable index information,and thus the index accuracy of data is greatly improved when the system RAM capacity is limited.The experimental results show that when a highly reliable antifuse field programmable gate array(FPGA)is used to implement a 1 TB SSD storage system and the system is required to retrieve pages,the index data only needs 5.9 Mb of RAM capacity overhead,the average write rate of the system is 118 Mbit/s,and the average index delay is 40.1μs.

关 键 词:深空探测 探测器设计 固态存储系统 日志型数据管理 索引精度 

分 类 号:TP333[自动化与计算机技术—计算机系统结构]

 

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