目前SSD技術發展越來越成熟，隨著容量逐漸的增加，而價格也越來低廉，比起一般Hard Disk讀取和寫入速度快出許多，因此對於未來儲存體的技術，想必會造成很大的影響。但SSD相對遭遇到壽命長短的問題，對於每個cell的寫入次數有限以及garbage collection overhead都相對需要受到考量。而現階段SSD產品皆為SATA介面，須經由Southbridge溝通，但對於SSD的存取速度而言，皆受到通道速度的限制。因此我們專注在目前最新技術的產品PCIE-SSD，透過Northbridge進行溝通，因此如何有效的改善PCIE-SSD之效能及壽命成為我們本篇碩論的研究發展方向。實驗結果顯示，我們成功改善效能以及壽命，提高整體性能，與對手相比，最佳性能改善了67.8％，而平均寫入最好時間是73％。

The SSD technology development is more and more mature which the read and write performance are better than hard disk. For this reason, the technology of storage device will caused a great impact in the future. Although SSD has a lot of advantages, it also encounter some problems that life-time and garbage collection overhead should be solved. In present of SSD devices on market are almost SATA interface which should connect through southbridge. For the SATA interface, it suffered from speed bottleneck of SSD which the access speed of the SSD are limit of the channel. The SATA 2.0 and SATA 3.0 interface of speed is 3 Gb/s and 6 Gb/s respective. If the next generation of SATA interface is become more fast, it can not exceed the PCIe interface (the PCIe 1.1x4 speed is 10Gb/s). One of PCIe interface advantage is that can provide 10Gb Ethernet equate bandwidth. The other advantage is that it has more than 1 of PCIe 1.1x16 (the PCIe 1.1x16 speed is 40Gb/s) or more than 2 of PCIe 1.1x8 (the PCIe 1.1x8 speed is 20Gb/s) in computer motherboard now. Our studies are focus on PCIe-SSD of the latest technology product which use the northbridge to get connection and can quickly deliver the data. It can let SSD product get more effective, but it still suffered the problem of NAND flash memory. Our work can improve the PCIe-SSD performance and lift-time more effective than others work. The experimental results display that we are successful to improve wear leveling and increase the overall performance which the best performance 67.8\% and the best average write response time is 73\% on trace file compare with the opponent.

[1] 4Gb, 8Gb, and 16Gb x8 NAND Flash Memory Features, Micron technology,2006.
[2] 32Gb, 64Gb, 128Gb, 256Gb Asynchronous/Synchronous NAND Features, Micron technology, 2010.
[3] S.-W. Lee, D.-J. Park, T.-S. Chung, D.-H. Lee, S. Park, and H.-J. Song, “A log buffer-based flash translation layer using fully-associative sector translation,” ACM Trans. Embed. Comput. Syst., vol. 6, Jul. 2007.
[4] Q. Wei, B. Gong, S. Pathak, B. Veeravalli, L. Zeng, and K. Okada, “Waftl: A workload adaptive flash translation layer with data partition,” in Proceedings of the 2011 IEEE 27th Symposium on Mass Storage Systems and Technologies, ser. MSST ’11. Washington, DC, USA: IEEE Computer Society, 2011,
pp. 1–12.
[5] S. Lee, D. Shin, Y.-J. Kim, and J. Kim, “Last: locality-aware sector translation for nand flash memory-based storage systems,” SIGOPS Oper. Syst. Rev., vol. 42, pp. 36–42, Oct. 2008.
[6] H. Cho, D. Shin, and Y. I. Eom, “Kast: K-associative sector translation for nand flash memory in real-time systems,” in Proceedings of the Conference
on Design, Automation and Test in Europe, ser. DATE ’09. 3001 Leuven,Belgium, Belgium: European Design and Automation Asso
[7] O. corporation,http://www.ocztechnology.com/tw/.ciation, 2009, pp.507–512.
[8] P. M. Chen, E. K. Lee, G. A. Gibson, R. H. Katz, and D. A. Patterson,“Raid: high-performance, reliable secondary storage,” ACM Comput. Surv.,vol. 26, pp. 145–185, Jun. 1994.
[9] F.-I. corporation, http://www.fusionio.com/.
[10] L. corporation, http://www.lsi.com/Pages/default.aspx.
[11] A. Gupta, Y. Kim, and B. Urgaonkar, “Dftl: a flash translation layer employing demand-based selective caching of page-level address mappings,” SIGPLAN Not., vol. 44, no. 3, pp. 229–240, Mar. 2009.
[12] L.-P. Chang, “A hybrid approach to nand-flash-based solid-state disks,” pp.
1337–1349, 2010.
[13] L.-P. Chang and C.-D. Du, “Design and implementation of an efficient wear-leveling algorithm for solid-state-disk micro-controllers,” pp. 1–36, 2009.
[14] UMassTraceRepository, “Oltp application i/o,”
http://traces.cs.umass.edu/index.php/Storage/Storage.