固態硬碟已經被預期當作下一世代的儲存裝置。由於固態硬碟同時利用許多快閃記憶體元件，因此有兩種因素將會影響到固態硬碟的反應速度 : 平行處理程度和回收實體區塊所需要的負擔程度。並且，在固態硬碟中，有兩種資料放置方式: 以頁面為單位作分散，可以有效增加平行處理的程度；以區塊作單位的分散，可以減少回收實體區塊所需要的負擔。因此這篇研究，提出在邏輯區塊中，資料的放置方式依照資料的特性而做擺放。讀取密集的邏輯區塊由於稀少引發回收實體區塊，所以採用頁面為單位作分散來增加平行度。寫入密集的邏輯區塊，採用區塊作單位的分散來減少回收實體區塊所需要的負擔。提出技術可以對每一個區塊找出最合適的資料放置方式，並且提出一個緩衝層和轉譯層的合作技術，來達到平行度。在我們實驗結果顯示，我們提出的方法能達到較好的適應性，並且考慮平行處理和回收負擔以獲得比較好的效率。

Solid-state drives (SSDs) have been expected as the next generation storage device. Owing to the SSD uses many NAND flash packages concurrently, there are two factors which effect the response time: the degree of parallelism and the GC (garbage collection) overhead. There are two kinds of page placement: page striping and block striping, which page striping is effective to increase the degree of parallelism and block striping benefits to reduce the GC overhead. Therefore, this paper proposed a Block-Level and Adaptive Striping (BLAS) policy that adaptive page placement depend on the access pattern in the block-level. Read-intensive logical blocks adopt page striping to increase the degree of parallelism and write-intensive logical blocks adopt block striping to reduce the GC overhead. The goal of BLAS is to find optimal per-block placement policy for any workload. Specially, a parallel write page in block-level buffer management is proposed for write-intensive data to satisfy the high degree of parallelism. Our experiments show the adaptiveness of BLAS under ten workloads. Parallelism and GC overhead both can be considered in BLAS to obtain better performance.

[1] Samsung Electronics, 1G x 8 Bit / 2G x 8 Bit NAND Flash Memor, http://www.datasheetcatalog.com/datasheets_pdf/K/9/W/A/K9WAG08U1M.shtml.2005
[2] 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. Embedded Computing Systems, vol. 6, no. 3, pp. 436-453, 2007.
[3] D. Jung, J.-U. Kang, H. Jo, J.-S. Kim, and J. Lee, “Superblock FTL: A superblock-based flash translation layer with a hybrid address translation scheme,” ACM Trans. Embedded Computing Systems, vol. 9, no. 4, pp. 1-41, 2010.
[4] A. Kawaguchi, S. Nishioka, and H. Motoda, “A Flash-Memory Based File System,” in Proc. 1995 USENIX Winter Technical Conf.,pp. 155-164. 1995
[5] M.-L. Chiang, P. C. H. Lee, and R.-C. Chang, “Cleaning Policies in Mobile Computers Using Flash Memory,” Journal of Systems and Software, vol. 48, no. 3, pp. 213-231, 1999.
[6] M. Chiang, P. Lee, and R. Chang, “Using data clustering to improve cleaning performance for flash memory,” Software-Practice and Experience, vol. 29, no. 3, pp. 267-290, 1999.
[7] L. P. Chang, and T. W. Kuo, “An Adaptive Striping Architecture for Flash Memory Storage Systems of Embedded Systems,” in Proc. IEEE Real-Time and Embedded Technology and Applications Symposium, Beijing China, 2002, pp. 187 - 196.
[8] B. Ha, H. Cho, and Y. I. Eom, “A study on the block fragmentation problem of ssd based on NAND flash memory,” in Proc., 2011, pp. 100.
[9] J. Shin, Z. Xia, N. Xu, R. Gao, X. Cai, S. Maeng, and F. Hsu, “FTL design exploration in reconfigurable high-performance SSD for server applications,” in Proc. 23rd Int. Conf. Supercomputing, 2009, pp. 338-349
[10] S. Lee, K. Fleming, J. Park, K. Ha, A. Caulfield, S. Swanson, and J. Kim, “BlueSSD: an open platform for cross-layer experiments for NAND Flash-based SSDs,” in Proc., 2010, pp.
[11] S. Out, “Design and analysis of flash translation layers for multi-channel NAND flash-based storage devices,” IEEE Transactions on Consumer Electronics, vol. 55, no. 3, pp. 1392-1400, 2009.
[12] S. Park, J. Park, S. Kim, and C. Weems, “A Pattern Adaptive NAND Flash Memory Storage Structure,” IEEE Transactions on Computers, 2010.
[13] J. Kang, J. 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, no. 9, pp. 644-658, 2007.
[14] H. Kim, and S. Ahn, “BPLRU: A buffer management scheme for improving random writes in flash storage,” in Proc. USENIX Conf. File and Storage Technologies, 2008, pp. 1-14.
[15] S. Lee, D. Shin, Y. Kim, and J. Kim, “LAST: locality-aware sector translation for NAND flash memory-based storage systems,” ACM SIGOPS Operating Systems Review, vol. 42, no. 6, pp. 36-42, 2008.
[16] S. Kang, S. Park, H. Jung, H. Shim, and J. Cha, “Performance trade-offs in using nvram write buffer for flash memory-based storage devices,” IEEE Transactions on Computers, vol. 58, no. 6, pp. 744-758, 2009.
[17] J. Hu, H. Jiang, L. Tian, and L. Xu, “PUD-LRU: An Erase-Efficient Write Buffer Management Algorithm for Flash Memory SSD,” in Proc. 2010 18th Annual IEEE/ACM International Symposium on Modeling, Analysis and 2010, pp.
[18] G. Wu, B. Eckart, and X. He, “BPAC: An Adaptive Write Buffer Management Scheme for Flash-based Solid State Drives,” 2010.
[19] S. K. Park, Y. Park, G. Shim, and K. H. Park, “CAVE: channel-aware buffer management scheme for solid state disk,” in Proc. 26th Annu. ACM Symp. on Applied Computing, TaiChung, Taiwan, 2011, pp. 346-353.
[20] Microsoft, SSD Extension for DiskSim Simulation Environment http://research.microsoft.com/en-us/downloads/b41019e2-1d2b-44d8-b512-ba35ab814cd4/
[21] Samsung Electronics, 2G x 8 Bit / 4G x 8 Bit / 8G x 8 Bit NAND Flash Memor, http://tec.icbuy.com/upload/database/2007-09/0709031723090581428001188811389.pdf
[22] Oracle, VirtualBox, http://www.virtualbox.org/
[23] blktrace(8)-Linux man page, http://linux.die.net/man/8/blktrace
[24] Microsoft, SQLIOSim, http://support.microsoft.com/kb/231619
[25] Storage Networking Industry Association, IO traces, http://www.snia.org/
[26] Taiwan university flash group, IO traces, http://newslab.csie.ntu.edu.tw/~flash/index.php?SelectedItem=Traces
[27] UMASS Amherst, OLTP IO traces, http://www.umass.edu/
[28]Microsoft, DiskMon, http://technet.microsoft.com/en-us/sysinternals/bb896646
[29] H. Kwon, E. Kim, J. Choi, D. Lee, and S. H. Noh, “Janus-FTL: finding the optimal point on the spectrum between page and block mapping schemes,” in Proc., 2010, pp. 169-178.