簡易檢索 / 詳目顯示

研究生: 林佳豪
Chia-Hao - Lin
論文名稱: 針對物件式儲存架構的使用者定義檔案系統
A User-Defined File System for Object Storages
指導教授: 吳晉賢
Chin-Hsien Wu
口試委員: 林淵翔
Yuan-Hsiang Lin
謝仁偉
Jen-Wei Hsieh
修丕承
Pi-Cheng Hsiu
學位類別: 碩士
Master
系所名稱: 電資學院 - 電子工程系
Department of Electronic and Computer Engineering
論文出版年: 2017
畢業學年度: 105
語文別: 中文
論文頁數: 85
中文關鍵詞: 物件式儲存軟體定義儲存檔案系統
外文關鍵詞: object storage, software-defined storage, file system
相關次數: 點閱:232下載:0
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

目前一些軟體定義儲存系統朝著非結構化資料、擴展性、富彈性、可靠性與虛擬化發展。其中如 Amazon S3 或 Google Cloud Storage 等這類的物件式儲存系統已經逐漸熱門普及,越來越多企業採用。然而,隨著使用者需求與物件式儲存種類的增加,物件式儲存的管理以及資料儲存的最佳化將變成是個重要的議題。在本論文中,我們提出一種稱為「針對物件式儲存架構的使用者定義檔案系統」的檔案系統架構。這個架構允許以使用者定義的儲存策略去管理物件式儲存,並且兼具物件式儲存與 POSIX 檔案系統的優點。


Today, some software-defined storage systems are developed for unstructured data, scalability, flexibility, reliability, and virtualization. In particular, an object-based storage system (e.g. Amazon S3, Google Cloud Storage) has been progressively popularized. More and more enterprises have integrated them into their storage systems. However, object storage management and data store optimization will become important problems when user requirements and different types of object storages are considered. In the thesis, we will propose a user-defined file system for object storages. The purpose of the thesis is to manage object storages with user-defined storage policies and combine the benefits of object storages with a POSIX file system.

論文摘要 I Abstract II 目錄 III 圖目錄 VII 表目錄 X 1 緒論 1 2 背景知識與相關研究 4 2.1 具象狀態轉移 4 2.1.1 總覽 4 2.1.2 客戶端-伺服端架構 4 2.1.3 無狀態 4 2.1.4 可快取 5 2.1.5 統一接口 5 2.1.6 分層系統 7 2.1.7 依程式碼腳本 7 2.1.8 小結 7 2.2 軟體定義儲存 8 2.2.1 總覽 8 2.2.2 可擴展的架構 10 2.2.3 常見的硬體 11 2.2.4 資源池 11 2.2.5 抽象化 11 2.2.6 自動化 11 2.2.7 可程式化 12 2.2.8 策略驅動 12 2.2.9 小結 12 2.3 物件式儲存 12 2.3.1 總覽 12 2.3.2 物件式儲存裝置 13 2.3.3 物件 14 2.3.4 物件式儲存檔案系統vs.傳統POSIX檔案系統 15 2.3.5 小結 17 2.4 使用者空間檔案系統 17 2.4.1 總覽 17 2.4.2 小結 18 2.5 其他研究 19 3 動機 21 4 針對物件式儲存架構的使用者定義檔案系統 22 4.1 總覽 22 4.2 物件式儲存 26 4.3 儲存驅動 27 4.3.1 檔案系統-FS 30 4.3.2 檔案元資料-FileMetadata 31 4.3.3 檔案-Files 33 4.3.4 檔案系統層級儲存驅動 34 4.4 安全的元資料儲存 36 4.4.1 物件式儲存元資料-StorageMetadata 37 4.4.2 檔案元資料-FileMetadata 38 4.4.3 檔案-Files 39 4.5 使用者定義的管理策略 41 4.5.1 中間層 42 4.5.2 儲存策略 43 4.5.3 FUSEI/O處理 44 4.6 溝通接口 47 4.6.1 I/O接口 47 4.6.2 控制接口 48 4.7 物件式儲存處理器 48 4.8 其他 48 4.9 小結 49 5 實驗結果與分析 50 5.1 實驗環境 50 5.2 不同運算效能下的效能量測 51 5.2.1 UDFS設定 52 5.2.2 檔案系統操作反應時間 52 5.2.3 不同大小的檔案與區塊讀寫效能 54 5.3 不同儲存系統下的效能量測 58 5.3.1 UDFS設定 58 5.3.2 實驗方法 59 5.3.3 實驗結果 60 5.4 使用者定義能力的驗證 65 5.4.1 UDFS設定 65 5.4.2 實驗方法 66 5.4.3 實驗結果 67 6 結論與後續工作 70 6.1 結論 70 6.2 後續工作 70 參考文獻 71

[1] VMware,“VMware.”,1999.[Online;accessed12-Oct- 2016].
[2] Qumranet, “KVM.”, 2012. [Online; accessed 12- Oct-2016].
[3] K. Fraser, S. Hand, and I. Pratt, “Xen.”, 2003. [Online; accessed 12-Oct-2016].
[4] S. Soltesz, H. Pötzl, M. E. Fiuczynski, A. Bavier, and L. Peterson, “Container-based operating system virtualization: a scalable, high-performance alternative to hyper- visors,” in ACM SIGOPS Operating Systems Review, vol. 41, pp. 275–287, ACM, 2007.
[5] S.Hykes,“Docker.” ,2013.[Online;accessed12-Oct- 2016].
[6] H. TECHNOLOGIES, “RAID 2.0+ Technical White Paper.”, 2014. [Online; accessed 12-Oct- 2016].
[7] S. A. Weil, S. A. Brandt, E. L. Miller, D. D. Long, and C. Maltzahn, “Ceph: A scalable, high-performance distributed file system,” in Proceedings of the 7th symposium on Operating systems design and implementation, pp. 307–320, USENIX Association, 2006.
[8] M. Mesnier, G. R. Ganger, and E. Riedel, “Object-based storage,” IEEE Communications Magazine, vol. 41, no. 8, pp. 84–90, 2003.
[9] Wikipedia, “Direct-attached storage.”, 2016. [Online; accessed 12-Oct-2016].
[10] Wikipedia, “Storage area network.”, 2016. [Online; accessed 12-Oct-2016].
[11] Wikipedia, “Network-attached storage.”, 2016. [Online; accessed 12-Oct-2016].
[12] D. Networks, “A beginner’s guide to next generation object storage.”, 2016. [Online; accessed 12-Oct-2016].
[13] Amazon, “Amazon S3.”, 2006. [Online; accessed 12-Oct-2016].
[14] Google, “Google Cloud Storage.”, 2010. [Online; accessed 12-Oct-2016].
[15] J. Arnold, OpenStack Swift: Using, Administering, and Developing for Swift Object Storage. ” O’Reilly Media, Inc.”, 2014.
[16] D. Beaver, S. Kumar, H. C. Li, J. Sobel, P. Vajgel, et al., “Finding a needle in haystack: Facebook’s photo storage.,” in OSDI, vol. 10, pp. 1–8, 2010.
[17] R. T. Fielding, Architectural styles and the design of network-based software architectures. PhD thesis, University of California, Irvine, 2000.
[18] C. Richardson, “Pattern: Microservices architecture,” Microservices. io. http://microservices. io/patterns/microservices. html [last accessed on February 17, 2015], 2014.
[19] Y.Jararweh,M.Al-Ayyoub,E.Benkhelifa,M.Vouk,A.Rindos,etal.,“Software defined cloud: Survey, system and evaluation,” Future Generation Computer Systems, vol. 58, pp. 56–74, 2016.
[20] F. Wu and G. Sun, “Software-defined storage,” Report. University of Minnesota, Minneapolis, 2013.
[21] N. McKeown, T. Anderson, H. Balakrishnan, G. Parulkar, L. Peterson, J. Rexford, S. Shenker, and J. Turner, “Openflow: enabling innovation in campus networks,” ACM SIGCOMM Computer Communication Review, vol. 38, no. 2, pp. 69–74, 2008.
[22] A. Darabseh, M. Al-Ayyoub, Y. Jararweh, E. Benkhelifa, M. Vouk, and A. Rindos, “Sdstorage: a software defined storage experimental framework,” in Cloud Engineering (IC2E), 2015 IEEE International Conference on, pp. 341–346, IEEE, 2015.
[23] E. Thereska, H. Ballani, G. O’Shea, T. Karagiannis, A. Rowstron, T. Talpey, R. Black, and T. Zhu, “Ioflow: a software-defined storage architecture,” in Proceedings of the Twenty-Fourth ACM Symposium on Operating Systems Principles, pp. 182–196, ACM, 2013.
[24] C. Lu, “SeaweedFS.”, 2012. [Online; accessed 12-Oct-2016].
[25] D. Hardt, “The oauth 2.0 authorization framework,” 2012.
[26] M. Szeredi, “Filesystem in userspace.”, 2005. [Online; accessed 12-Oct-2016].
[27] M. Buus, “fuse-bindings.”,
2014. [Online; accessed 12-Oct-2016].
[28] J. Kaylor, K. Läufer, and G. K. Thiruvathukal, “Restfs: resources and services are filesystems, too,” in Proceedings of the Second International Workshop on RESTful Design, pp. 39–46, ACM, 2011.
[29] K.-L. Chen and C.-H. Wu, “A deduplication-aware scalable storage system for mo- bile devices,” 2014.
[30] C.-K.Kang,Y.-J.Cai,C.-H.Wu,andP.-C.Hsiu,“A hybrid storage access framework for high-performance virtual machines,” ACM Transactions on Embedded Computing Systems (TECS), vol. 13, no. 5s, p. 157, 2014.
[31] S. Sinclair and B. Lundell, “Object Storage Trends.”, 2015. [On- line; accessed 12-Oct-2016].
[32] W. Vogels, “Eventually consistent,” Communications of the ACM, vol. 52, no. 1, pp. 40–44, 2009.
[33] K. Chodorow, MongoDB: the definitive guide. ” O’Reilly Media, Inc.”, 2013.
[34] Amazon, “Object Key and Metadata.”, 2016. [Online; accessed 12-Oct-2016].
[35] J. Dean and S. Ghemawat, “LevelDB.”, 2011. [Online; accessed 12-Oct-2016].
[36] K. Tseng, “LevelDB Benchmarks.”, 2011. [Online; accessed 12-Oct-2016].
[37] D.S.Roselli,J.R.Lorch,T.E.Anderson,et al.,“A comparison of file system workloads.,” in USENIX annual technical conference, general track, pp. 41–54, 2000.
[38] Node.js, “Node.js v6.8.1 Documentation - Stream.”, 2015. [Online; accessed 12-Oct-2016].
[39] H. Yaapa and R. McKinney, “Writing middleware for use in Express apps.”, 2015. [Online; accessed 12-Oct-2016].

無法下載圖示 全文公開日期 2022/01/20 (校內網路)
全文公開日期 本全文未授權公開 (校外網路)
全文公開日期 本全文未授權公開 (國家圖書館:臺灣博碩士論文系統)
QR CODE