面向SSD性能提升与寿命延长的缓存算法设计
面向SSD性能提升与寿命延长的缓存算法设计(任务书,开题报告,论文14000字)
摘要
传统的磁盘存储器主要是机械硬盘，机械硬盘的寿命主要由工作时间决定，其磁盘的擦除次数对寿命的影响不大，故传统的磁盘管理算法不用考虑磁盘的擦除次数，但是固态硬盘的寿命受磁盘擦除次数的影响，故传统的磁盘管理算法不能直接搬移到固态盘管理算法上，需要重新设计。目前常用的方法是给固态硬盘加一层缓存，这样可以增大数据吞吐量，降低响应时延，减少因写请求导致的擦除操作，既提高了存储器的性能又延长了寿命。其中缓存算法主要采用了最近最少使用算法。通过建模仿真，定量分析缓存算法对固态盘性能和寿命的影响发现，随着缓存的增大，固态盘的擦除次数减少、单位时间数据吞吐量增大、响应时间减少，故缓存有效提高固态盘的性能和延长固态盘的寿命。
关键词：固态硬盘；缓存算法；寿命；性能
Abstract
The traditional disk storage is mainly a mechanical hard disk. The life of the mechanical hard disk is determined by the working time. The disk erasure count has little effect on its life. Therefore, the traditional disk management algorithm doesn’t attach importance to the disk erasure count.However,the erasure affects the life of solid state diskgreatly. Therefore, the traditional disk management algorithms cannot be directly applied to the SSD management algorithm and need to be redesigned. Currently, a commonly used method is to add a layer of cache to the solid state disk, which can increase the data throughput, reduce the response delay, and reduce the erase operation caused by the write request, which not only improves the performance of the flash but also prolongs the its life. The cache algorithm mainly uses the least recently used algorithm. By modeling and simulating, quantitative analysis of the impact of the cache algorithm on the performance and life span of the solid state disk found that with the increase of the cache, the erasing frequency of the solid state disk is reduced, the throughput per unit time of data is increased, and the response time is reduced. Therefore, the cache has effectively increased solidstate disk performance and extended SSD’s life.
Keywords: solid state disk; cache algorithm; lifetime; performance
目录
摘要 1
Abstract 2
第1章绪论 1
1.1背景介绍 1
1.2固态盘的特性 2
1.3主流的缓存技术 3
1.3.1最佳置换算法（Optimal） 3
1.3.2先进先出页面置换算法（FIFO） 4
1.3.3最近最久未使用置换算法（LRU） 5
1.4国内外研究现状 6
1.5研究意义和内容 6
第2章缓存算法的设计 8
2.1实际场景的需求分析 8
2.2内存的页面置换算法的设计 9
2.3缓存的读写模块设计 10
2.4本章小结 11
第3章缓存算法的实现 13
3.1请求数据的结构 13
3.2内存的页面置换算法的实现 13
3.3闪存读写模块的实现 14
3.3.1 4K对齐的实现 14
3.3.2 缓存页的拼接 14
3.3.3 缓存页与闪存的数据一致性 15
3.4本章小结 15
第4章缓存算法的性能测试与分析 16
4.1测试平台 16
4.2测试数据集来源 16
4.3性能测试 16
4.4测试结果分析 18
4.5本章小结 18
第5章总结和展望 19
5.1全文总结 19
5.2未来的展望 19
参考文献 20
致谢 21 [资料来源：Doc163.com]