Flash-based的儲存裝置由於存取速度(特別是隨機讀取)的速度比起傳統硬碟還要快上很多，其中由NAND flash組成的SSD近幾年來在很多應用領域都已經漸漸取代傳統機械式硬碟。SSD中扮演效能關鍵角色的就是FTL，其主要負責(1)邏輯位址與實體位址轉換(2)空間回收的功能，又由於FLASH本身的特性，還需要維持(3)平均抹除次數的平衡，以延長SSD的使用壽命。本篇論文著重在存取速度的效能提升，大部分的研究藉由修改寫入的機制來減少系統的回應時間，對於讀取部分的加速通常僅利用workload的時間區域性、空間區域性以及利如DRAM這樣的快取來執行。我們的想法是：實際上某些workload其應用較傾向於讀取為主，可能出現某些常用資料很常讀取而很少更新，造成整體讀取次數遠大於寫入次數的情形。此外有些FTL設計之初並沒有加入平行化處理的考量，並且現在的SSD大都採用multi-channel的架構，且支援進階指令集。加入這些考量可以更加提升其效能。本篇論文主要以現在普遍的multi-channel架構為目標硬體，利用平行化以及進階指令集的概念，在不增加過多負載的情況下，修改現有FTL的相關機制，針對以讀取為主的workload增進其存取的效能。

The access speed of flash-based storage devices is much faster than traditional hard disk (especially random reads), the SSD, which composed by NAND flash, is more popular in the few years. In the SSD, FTL(Flash Translation Layer)plays a key role for performance, which is mainly responsible for (1)Translate the logical address to the physical address, (2)Garbage collection, and based on the characteristics of the FLASH,it's also need to maintain (3)Wear-leveling, to extend the lifetime of the SSD. This paper focus on Improvement of access speed(more accurate, read speed). Most of related researchers analyze the write mechanism, modify it to decrease the response time, and use cache memory to service read requests. But in our idea, some workloads' pattern show different characteristics for read. Some data is often read and rarely updated. In the overall, number of reads is much larger than writes. Besides, the design of some FTL and not adding parallel processing considerations, and most of the SSD using multi-channel architecture and support some advanced commands now. By adding the considerations, we can more improve the performance. This paper uses the common multi-channel architecture for the target hardware, and considers the concept of parallel and advanced commands. Without increasing too much overhead to modify the the existing FTL mechanism to enhance the performance of read-domain workloads.

[1] A. Gupta, Y. Kim, and B. Urgaonkar, \Dftl: a
ash translation layer
employing demand-based selective caching of page-level address mappings,"
SIGPLAN Not., vol. 44, no. 3, pp. 229{240, Mar. 2009. [Online]. Available:
http://doi.acm.org/10.1145/1508284.1508271
[2] S.-W. Lee, D.-J. Park, T.-S. Chung, D.-H. Lee, S. Park, and H.-J. Song,
\A log bu er-based
ash translation layer using fully-associative sector
translation," ACM Trans. Embed. Comput. Syst., vol. 6, no. 3, Jul. 2007.
[Online]. Available: http://doi.acm.org/10.1145/1275986.1275990
[3] S. Lee, D. Shin, Y.-J. Kim, and J. Kim, \Last: locality-aware sector
translation for nand
ash memory-based storage systems," SIGOPS Oper.
Syst. Rev., vol. 42, no. 6, pp. 36{42, Oct. 2008. [Online]. Available:
http://doi.acm.org/10.1145/1453775.1453783
[4] D. Jung, J.-U. Kang, H. Jo, J.-S. Kim, and J. Lee, \Superblock ftl: A
superblock-based
ash translation layer with a hybrid address translation
scheme," ACM Trans. Embed. Comput. Syst., vol. 9, no. 4, pp. 40:1{40:41,
Apr. 2010. [Online]. Available: http://doi.acm.org/10.1145/1721695.1721706
[5] F. Chen, T. Luo, and X. Zhang, \CAFTL: A Content-Aware Flash Translation
Layer Enhancing the Lifespan of Flash Memory Based Solid State
Drives," in Proceedings of the 9th USENIX Conference on File and Stroage
Technologies, ser. FAST'11, 2011, pp. 77{90.
[6] G. Shim, Y. Park, and K. H. Park, \A hybrid
ash translation layer with
adaptive merge for ssds," ACM Transactions on Storage, vol. 6, no. 4, p. 15,
May 2011.
[7] R. Bux, W.IBM Res. Zurich, S. X.-Y. Hu, I. Iliadis, and R. Haas,
\Scheduling in
ash-based solid-state drives - performance modeling and
optimization," SIGPLAN Not., vol. 44, no. 3, pp. 229{240, Mar. 2009.
[Online]. Available: http://doi.acm.org/10.1145/1508284.1508271
[8] F. Chen, T. Luo, and X. Zhang, \Essential Roles of Exploiting Internal Parallelism
of Flash Memory based Solid State Drives in High-Speed Data Processing,"
in IEEE 17th International Symposium on High Performance Computer
Architecture (HPCA), Feb 2011, pp. 266{277.
[9] Y. Hu, H. Jiang, D. Feng, L. Tian, H. Luo, and C. Ren, \Exploring and
Exploiting the Multi-level Parallelism Inside SSDs for Improved Performance
and Endurance," IEEE Transactions on Computers, vol. 58, no. 12, p. 1, Mar
2012.
[10] S. yeong Park, E. Seo, J.-Y. Shin, S. Maeng, and J. Lee, \Exploiting Internal
Parallelism of Flash-based SSDs," IEEE Computer Architecture Letters,
vol. 9, no. 1, pp. 9{12, Jan 2010.
[11] S.-H. Park, S.-H. Ha, K. Bang, and E.-Y. Chung, \Design and Analysis
of Flash Translation Layers for Multi-Channel NAND Flash-based Storage
Devices," IEEE Transaction on Consumer Electronics, vol. 55, no. 3, pp.
1392{1400, August 2009.
[12] S. K. Park, Y. Park, G. Shim, and K. H. Park, \CAVE: Channel-Aware
Bu er Management Scheme for Solid State Disk," in ACM Symposium on
Applied Computing, 2011, pp. 346{353.
[13] J.-U. Kang, J.-S. Kim, C. Park, H. Park, and J. Lee, \A Multi-Channel Architecture
for High-Performance NAND Flash-Based Storage System," Journal
of Systems Architecture, vol. 53, pp. 644{658, September 2007.
[14] L.-P. Chang, \A Hybrid Approach to NAND-Flash-Based Solid-State Disks,"
IEEE Transactions on Computers, vol. 59, no. 10, pp. 1337{1349, October
2010.
[15] \UMass Trace Repository." [Online]. Available: http://traces.cs.umass.edu
[16] Micron, \16Gb, 32Gb, 64Gb, 128Gb Asynchronous/Synchronous NAND Features,"
Micron, Tech. Rep., 2009.
[17] ||, \NAND Flash Memory MT29F4G08AAA, MT29F8G08BAA,
MT29F8G08DAA, MT29F16G08FAA," Micron, Tech. Rep., 2006.
[18] ||, \TN-29-25: Improving Performance Using Two-Plane Commands Introduction,"
Micron, Tech. Rep., 2007.
[19] Samsung, \K9GAG08B0M (2G x 8 Bit NAND Flash Memory) Datasheet,"
Samsung, Tech. Rep., 2006.
[20] ||, \NAND Flash Memory K9GAG08B0M, K9GAG08U0M,
K9LBG08U1M," Samsung, Tech. Rep., 2007.
[21] MIRA, \3.3V 16Gb MLC NAND Flash Memory Speci cation and Technical
Notes," MIRA, Tech. Rep., 2011