星载大容量固态存储器快速可靠启动算法设计

发布时间：2018-10-18 11:50

【摘要】：为解决星载SSR启动缓慢,可靠性不足所带来的无法满足复杂、灵活的任务需求问题,研究星载大容量固态存储器(SSR)的索引建立机制,分析传统方案的优缺点并结合在轨运控需求、工作模式等特点,提出一种适合基于NAND FLASH的星载大容量固态存储器的快速启动算法.原有文件系统存储索引表来自于NAND FLASH空余区,算法增加新的保留区设计及相应的启动过程,避免重启时扫描空余区重新建立索引.针对空间环境单粒子效应带来的存储错误,NAND FLASH保留区的索引信息采取ECC编码、冗余备份、分区存储等可靠性措施,提高索引表的可靠性.本文介绍了应用保留区的启动工作机理以及不同模式下的更新方式,阐述了系统在功能性重启和故障性重启下的扫描方式,并建立数学模型分析算法的有效性,最后在使用欧比特NAND FLASH搭建的测试平台上进行验证.算法功能性重启索引建立耗时5.643 ms,故障重启索引建立73.985 ms,而传统算法重启索引建立50.37 s.实验结果表明,本算法显著减少了系统启动耗时.
[Abstract]:In order to solve the problem that spaceborne SSR can not meet the complex and flexible task requirements caused by slow startup and insufficient reliability, the indexing mechanism of (SSR) with large capacity on spaceborne is studied. By analyzing the advantages and disadvantages of the traditional scheme and combining the characteristics of on-orbit operation control and operation mode, a fast start algorithm for spaceborne large capacity solid state memory based on NAND FLASH is proposed. The original file system storage index table comes from the NAND FLASH spare area. The algorithm adds a new reserved area design and the corresponding startup process to avoid scanning the spare area to re-establish the index during restart. In order to improve the reliability of index table, ECC coding, redundant backup and partition storage are adopted for the index information of, NAND FLASH reserved area caused by single particle effect in space environment. This paper introduces the startup mechanism of the reserved area and the renewal mode under different modes, expounds the scanning mode of the system under the functional restart and the fault restart, and establishes the validity of the mathematical model analysis algorithm. Finally, it is verified on the test platform built by NAND FLASH. It takes 5.643 ms, to establish the functional restart index and 73.985 ms, for the traditional algorithm to restart the index. Experimental results show that the proposed algorithm can significantly reduce the system startup time.
【作者单位】：中国科学院空间科学与应用研究中心;中国科学院大学;
【基金】：中国科学院战略性先导科技专项资助(XDA04060300)
【分类号】：TP333

【参考文献】