跳到主要內容

臺灣博碩士論文加值系統

(44.200.27.215) 您好!臺灣時間:2024/04/20 08:11
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:林孟宣
研究生(外文):Meng-Shiuan Lin
論文名稱:具備負載感知之多層級儲存裝置的資料快取方法
論文名稱(外文):A Load-aware SSD Caching Scheme for Multi-tiered Storage Systems
指導教授:張軒彬張軒彬引用關係
口試委員:張大緯焦信達
口試日期:2017-07-27
學位類別:碩士
校院名稱:國立中興大學
系所名稱:資訊工程學系所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2017
畢業學年度:105
語文別:中文
論文頁數:50
中文關鍵詞:多層級儲存裝置資料快取機制
外文關鍵詞:Multi-tiered Storage SystemSSD Caching
相關次數:
  • 被引用被引用:0
  • 點閱點閱:196
  • 評分評分:
  • 下載下載:5
  • 收藏至我的研究室書目清單書目收藏:0
近年來,出現了快閃記憶體所組成的固態硬碟,雖然固態硬碟具有高存取效能,但在單位容量的成本上,與傳統硬碟相比,還是相當高昂。因此,產生了結合固態硬碟與傳統硬碟的多層級儲存裝置。在效能與單位容量價格上取得平衡。利用高效能的固態硬碟作為傳統硬碟的快取,以提升I/O效能,同時利用大容量的傳統硬碟獲取更大的儲存空間。

先前有研究提出具備負載平衡之混合式儲存裝置的資料快取方法”Load Balance Prize Caching”,藉由平衡不同裝置的負載量,使儲存裝置間的服務時間盡量相等。本篇論文改善Load Balance Prize Caching的方法,首先,除了考量不同儲存裝置的存取時間外,我們也考量固態硬碟在執行寫入動作時,所可能引發Garbage Collection之成本;除此之外,本論文也會計算request產生的Queuing Delay。最後,當發生失衡情形時,也設計request重新導向機制,重新導向適合的request至負擔較輕的儲存裝置,使儲存裝置間的負載量能再平衡。

本論文使用基於Disksim所開發之混合式硬碟模擬器,探討在實作上所遭遇之議題。根據模擬結果可以證明設計之快取策略,可以有效降低request的反應時間,提升整體系統效能。 
In recent years, there have been NAND flash-based Solid-State Drives (SSDs). Although SSDs have high access performance, but, compared with traditional hard disk drives (HDDs), SSDs are much more expensive. Consequently, many researchers have proposed a multi-tiered heterogeneous storage architecture that integrates SSDs with HDDs, so as to trade-off between the cost and performance. By using the SSDs as the cache of HDDs, the I/O performance can be improved while exploiting the capacity of HDDs to obtain a large storage space.

Previously, researchers had proposed a SSD data caching scheme, called Load Balance Prize Caching (LBPC). LBPC takes the loadings of different storage devices into consideration and balances the loadings between HDDs and SSDs. In this thesis, we enhance the LBPC scheme. First, we consider the cost of garbage collection caused by the writes in SSDs. Besides, we also take the queuing delay on the storage device into account. Finally, when the loadings of the SSDs and HDDs are imbalance, we redirect the requests from the heavily loaded device to the lighter one, so that the loadings of the storage devices can be rebalanced again.

On the basis of DiskSim simulator, we implement our proposed cache policy onto a multi-tiered storage simulator. From the experimental results, our proposed scheme can reduce the average response time and improve the overall performance.
第一章 緒論 1
1.1簡介 1
1.2研究動機 2
1.3貢獻 3
1.4論文架構 4
第二章 背景知識與相關研究 5
2.1 Optimal Partition Flash Cache Layer 5
2.1.1 OP-FCL簡介 5
2.1.2 OP-FCL架構 6
2.1.3存取成本計算模型 9
2.1.4 OP-FCL運作過程 11
2.2 Prize Caching 12
2.2.1系統架構 12
2.2.2模組介紹 13
2.2.3快取策略 14
2.2.4運作流程 16
2.3 Load Balance Prize Caching 16
2.3.1系統架構 17
2.3.2模組介紹 18
2.3.3 Random Threshold調整方式 19
2.3.4運作流程 22
2.3.5 Load Balance Prize Caching的潛在問題 23
第三章 設計 24
3.1 Load-aware SSD Caching機制 24
3.1.1 Queuing Delay計算 24
3.1.2 SSD Write Overhead 28
3.1.3 Post-processing 28
3.2系統架構 29
3.3模組介紹 30
3.4 Random Threshold計算 31
3.5 Request重新導向 36
3.6運作流程 38
第四章 實驗結果與討論 39
4.1實驗環境 39
4.2實驗結果 41
4.2.1 Random Threshold 41
4.2.2不同Delta值之效能比較 42
4.2.3動態調整Random Threshold的效能比較 43
第五章 結論與未來工作 48
參考文獻 49
[1] T. Kgil and T. Mudge, "FlashCache: A NAND Flash Memory File Cache for Low Power Web Servers," in Proceeding CASES '06 Proceedings of the 2006 international conference on Compilers, architecture and synthesis for embedded systems, 2006.
[2] Y. Kim and J. Kim, "DAC: A Device-Aware Cache Management Algorithm for Heterogeneous Mobile Storage Systems," IEICE TRANSACTIONS on Information and Systems, VOL.E91-D, NO.12, 2008.
[3] Y. Kim and J. Kim, "Device-aware cache replacement algorithm for heterogeneous mobile storage devices," in ICESS '07 Proceedings of the 3rd international conference on Embedded Software and Systems, 2007.
[4] Y.-J. Kim and J.-H. Kim, "ARC-H: Adaptive replacement cache management for heterogeneous storage devices," Journal of Systems Architecture Volume 58, Issue 2, 2012.
[5] Y. Kim, A. Gupta, B. Urgaonkar, P. Berman and A. Sivasubramaniam, "HybridStore: A Cost-Efficient, High-Performance Storage System Combining SSDs and HDDs," in IEEE 19th Annual International Symposium on Modelling, Analysis, and Simulation of Computer and Telecommunication Systems, 2011.
[6] S.-Y. Liao, “A Data Caching Scheme for Hybrid Hard Drives,” 2010.
[7] "Intel Optane Memory," [Online]. Available: https://www.intel.com/content/www/us/en/architecture-and-technology/optane-memory.html.
[8] "Disksim," [Online]. Available: http://www.pdl.cmu.edu/DiskSim/.
[9] J.-C. Luo, “A Load-Balancing Caching Scheme for Hybrid Storage,” 2013.
[10] "MAXTOR ATLAS10K4 146G, 10000rpm SCSI," [Online]. Available: http://ovahldy.blogspot.com/2011/03/recent-diskmodels-for-disksim.htm.
[11] "PCMark7," [Online]. Available: http://www.futuremark.com/benchmarks/pcmark.
[12] Y. Liu, J. Huang, C. Xie and Q. Cao, "RAF: A Random Access First Cache Management to Improve SSD-based Disk Cache," in Fifth IEEE Internatinoal Conference on Networking, Architecture, and Storage, 2010.
[13] Y. Oh, J. Choi, D. Lee and S. H. Noh, "Caching less for better performance: Balancing cache size and update cost of flash memory cache in hybrid storage systems," in 10th USENIX Conference on File and Storage Technologies, 2012.
連結至畢業學校之論文網頁點我開啟連結
註: 此連結為研究生畢業學校所提供,不一定有電子全文可供下載,若連結有誤,請點選上方之〝勘誤回報〞功能,我們會盡快修正,謝謝!
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top