在數位化的時代，人們使用數位化的方式來儲存它們的資料在本地
或遠端的儲存設備上，隨著對於儲存要求的快速成長，傳統的儲存技術
開始不敷使用，因此，有著集中控制器的軟體定義儲存對於處理這些資
料要求而言，變成一個不錯的選擇，雖然目前有些研究和服務已經開始
利用軟體定義儲存所帶來的一些好處，但在這些研究和服務中，對於如
何建置多層資料容錯機制來去滿足不同階層的容錯需求上較少被提到，
除此之外，運用兩個或多個資料容錯機制在同一個硬碟組上會有一些問
題，且效能不好，為了去解決以上這些情況，本研究配合一些節能的方
法來去使多層資料容錯機制能夠建置在單一硬碟組中，一系列的實驗顯
示，跟原有架構相比，我們提出的方法能夠有效地降低儲存系統的能源
消耗。

In the era of digitalization, people store their data digitally on local or remote storage. With the rapid growth of storage demands, traditional storage techniques are considered inefficient. Therefore, Software Defined Storage (SDS) become a viable option since it includes a centralized controller to process data requirements. Although studies and services has been proposed and implemented to exploit the benefit bought by SDS, there is little discussion on how to enable multilevel data fault tolerance on SDS to satisfy different level of fault tolerance requirement. In addition, applying two or more data fault tolerance mechanisms on a same disk group could be problematic and not energy efficient. To address above issues, this study enables multilevel data fault tolerance on a single disk group with energy-efficient considerations. A series of experiments show that the proposed scheme could reduce the storage system energy consumption significantly when compared with the original architecture.

Abstract iii
Chapter 1. Introduction 1
Chapter 2. Background and Motivation 6
2.1 Software-Defined Storage 6
2.2 Data Fault-Tolerance Mechanism 8
2.3 Motivation 10
Chapter 3. Energy-Efficient Multilevel Data Fault Tolerance Design 14
3.1 Overview 14
3.2 Multilevel Data Fault Tolerance 17
3.3 Cache Policy – Hot Data Identification 20
3.4 Storage Volume Adjustment Policy 22
3.5 Energy-Efficient Data Placement Strategy 24
Chapter 4. Performance Evaluation 27
Chapter 5. Conclusion 32
References 34

[1] Ibm storwize family @ONLINE, http://www-03.ibm.com/systems/
storage/storwize/index.html.
[2] Nexentastor @ONLINE, https://nexenta.com/products/
nexentastor.
[3] The fundamentals of software-defined storage simplicity at scale for cloud architectures. Technical report, Coraid Inc, 2015. Report.
[4] A. Alba, G. Alatorre, C. Bolik, A. Corrao, T. Clark, S. Gopisetty, and et al. Efficient and agile storage management in software defined environments. IBM Journal of Research and Development, 2014.
[5] W. C. Arnold, D. M. Chess, M. V. Devarakonda, A. Segal, and I. N. Whalley. Method for policy-based, autonomically allocated storage, Jan 2009.
[6] B. Calder, J. Wang, A. Ogus, N. Nilakantan, A. Skjolsvold, S. McKelvie, Y. Xu, S. Srivastav, J. Wu, H. Simitci, J. Haridas, C. Uddaraju, H. Khatri, A. Edwards, V. Bedekar, S. Mainali, R. Abbasi, A. Agarwal, M. F. ul Haq, M. I. ul Haq, D. Bhardwaj, S. Dayanand, A. Adusumilli, M. McNett, S. Sankaran, K. Manivannan, and L. Rigas. Windows azure storage: a highly available cloud storage service with strong consistency. Proceedings of the Twenty-Third ACM Symposium on Operating Systems Principles (SOSP 11, pages 143–157, 2011.
[7] Kuo Sheng Deng, Chin Feng Lee, Jerry Chou, Yi Chen Shih, Shang Hao Chuang, and Po Hsuan Wu. pnfs-based software-defined storage for information lifecycle management. 2015 International Conference on Cloud Computing and Big Data (CCBD), pages 89–92, Nov 2015.
[8] M. Devarakonda, D. Chess, I. Whalley, A. Segal, P. Goyal, A.Sachedina, and et al. Policy-based autonomic storage allocation. Self-Managing Distributed Systems, 2003.
[9] N. Devireddy and X. Chen. Policy based storage configuration, Feb 2003.

[10] G.-L. Feng, R. H. Deng, F. Bao, and J.-C. Shen. New efficient mds array codes for raid part ii: Rabin-like codes for tolerating multiple (greater than or equal to 4) disk failures. IEEE Transactions on Computers, 54(12):1473–1483, Dec 2005.
[11] FenggangWu and G. Sun. Software-defined storage. Technical report, University of Minnesota, Dec 2013. Report.
[12] D. Ford, F. Labelle, F. I. Popovici, M. Stokely, V.-A. Truong, L. Barroso, C. Grimes, and S. Quinlan. Availability in globally distributed storage systems. 9th USENIX Symposium on Operating Systems Design and Implementation (OSDI 10), 2010.
[13] J. L. Hafner. Weaver codes: Highly fault tolerant erasure codes for storage systems. 3rd USENIX Conference on File and Storage Technologies (FAST05), pages 211–224, 2005.
[14] C. Hollis. Introducing emc vipr: A breathtaking approach to software defined storage @ONLINE, http://chucksblog.typepad.
com/chucks_blog/2013/05/introducing-emc-vipr-a-breathtaking-approach-to-software-de fined-storage. html, 2013.
[15] C. Huang, H. Simitci, Y. Xu, A. Ogus, B. Calder, P. Gopalan, J. Li, and S. Yekhanin. Erasure coding in windows azure storage. 2012 USENIX Annual Technical Conference (ATC12), pages 15–26, 2012.
[16] G. Kandiraju, H. Franke, M. Williams, M. Steinder, and S. Black. Software defined infrastructures. IBM Journal of Research and Development, 58:1–13, 2014.
[17] C. Li, B. Brech, S. Crowder, D. Dias, H. Franke, M. Hogstrom, and et al. Software defined environments: An introduction. IBM Journal of Research and Development, 2014.
[18] H. C. Lim, S. Babu, and J. S. Chase. Automated control for elastic storage. Proceedings of the 7th international conference on Autonomic computing, pages 1–10, 2010.
[19] Dushyanth Narayanan, Austin Donnelly, and Antony Rowstron. Write off-loading: Practical power management for enterprise storage. ACM Transactions on Storage (TOS), 2008.
[20] K. Palanivel and B. Li. Anatomy of software defined storage challenges and new solutions to handle metadata. Technical report, University of Minnesota, Aug 2013. Report.


[21] K. V. Rashmi, N. B. Shah, D. Gu, H. Kuang, D. Borthakur, and K. Ramchandran. A solution to the network challenges of data recovery in erasure-coded distributed storage systems: A study on the facebook warehouse cluster. 5th USENIX Workshop on Hot Topics in Storage and File Systems (HotStorage 13), 2013.
[22] Seagate. Product manual barracuda 7200.11 serial ata @ONLINE,http://www.seagate.com/staticfiles/support/disc/manuals/desktop/Barracuda%207200.11/100507013e.pdf, 2008.
[23] S. Seshadri, P. H. Muench, L. Chiu, I. Koltsidas, N. Ioannou, R. Haas, Y. Liu, M. Mei, and S. Blinick. Software defined just-in-time caching in an enterprise storage system. IBM Journal of Research and Development, 58, Apr 2014.
[24] H. L. Truong and S. Dustdar. Principles of software-defined elastic systems for big data analytics. Proceedings of the 2014 IEEE International Conference on Cloud Engineering, 2014.
[25] S. Uttamchandani, K. Voruganti, S. Srinivasan, J. Palmer, and D. Pease. Polus: Growing storage qos management beyond a 4-year old kid. Proceedings of the 3rd USENIX Conference on File and Storage Technologies, pages 31–44, 2014.
[26] A. Verma, R. Koller, L. Useche, and R. Rangaswami. Srcmap: energy proportional storage using dynamic consolidation. 8th USENIX conference on File and storage technologies(FAST10), 2010.