( 您好!臺灣時間:2024/02/24 14:32
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::


研究生(外文):Chih-Hung Chou
論文名稱(外文):Improving Journaling Filesystem Performance on a Virtual Machine Monitor
指導教授(外文):Wei-Kuan Shih
外文關鍵詞:Virtual machine monitorJournaling filesystemXenExt3
  • 被引用被引用:0
  • 點閱點閱:254
  • 評分評分:
  • 下載下載:42
  • 收藏至我的研究室書目清單書目收藏:3
日誌式檔案系統使用 metadata journaling 的技術來增加檔案系統的效能及可靠性。然而其代價是需要額外的磁碟空間來儲存檔案系統更新日誌,此外記錄日誌所引起的磁碟活動也會造成效能下降。在虛擬機器架構下使用日誌式檔案系統時,由於前述虛擬機器架構上的限制使得額外的磁碟空間及日誌磁碟活動所造成的負擔加重。
在本論文中我們提出透過修改日誌式檔案系統與 Virtual Machine Monitor (VMM) 使其合作來減輕前述問題。我們利用在 VMM 中記錄檔案資料記憶體的相關資訊並且配合 VMM 來保護這些檔案資料,避免檔案在尚未寫回磁碟前因程式錯誤而損壞。透過這個機制,我們消除了需要額外將檔案更新資料寫回的日誌磁碟空間的需求而減輕了前述問題。根據實驗結果顯示,我們所提出的方法較原本日誌式檔案系統有約 30% 到 150% 的效能改進,而檔案系統修復的速度則快2到6倍。
Because virtual machine monitor (VMM) provides a secure, reliable and fully isolated virtual execution environment, it offers a novel solution to improve security, reliability and increase system utilization. However, because every virtual machine operates independently and utilizes individual disk partitions, disk seek time can easily increase when accessing different disk partitions. Further, VMM checks every I/O access of the virtual machine to ensure that no one can corrupt the system. Thus, it results in higher I/O access overhead under virtual machine architecture then under traditional operating system. So, it is important to adapt the filesystem to the virtual execution environment to improve performance.
Journaling filesystems adopt metadata journaling technique to improve filesystem performance and consistency. However, the cost brings by metadata journaling is extra disk space and performance degradation caused by frequent journal data flushing. In addition, the problem is intensified due to the limitations of virtual machine architecture as discussed above.
In this thesis, we propose a framework that modifies both the journaling filesystem and the VMM to alleviate the overhead. We only log necessary information of file buffer caches in the VMM and protect them from being tainted via the use of VMM. By using this mechanism we eliminate the need for extra disk space while still retaining same filesystem consistency. Experimental data suggests our method improves journaling filesystem performance by 30% to 150% and speeds up the filesystem recovery process by 2 to 6 times faster.
摘要 i
Abstract ii
致謝 iii
章節目錄 iv
圖表目錄 vi
第一章 緒論 1
1.1 動機 1
1.2 論文組織 3
第二章 相關背景研究 4
2.1 Journaling Filesystem 4
2.1.1 Importance of Filesystem Consistency 4
2.1.2 Metadata/Data Journaling Technique 5
2.1.3 Linux Ext3 Filesystem 5
2.2 Virtual Machine 7
第三章 相關研究 10
3.1 Soft Update 10
3.2 Metadata Snapshotting 10
3.3 Log-structured Filesystem(LFS) 11
3.4 NVRAM 12
3.5 RAM I/O(Rio) 12
3.6 Xen FS 13
第四章 設計與實作 15
4.1 設計目標 15
4.2 系統設計 15
4.2.1 Protection Mechanism 16
4.2.2 Recovery Mechanism 18
4.2.3 Unified Journaling API 19
4.3 系統架構 19
4.3.1 Journaling Component 20
4.3.2 VMM Component 21
4.4 系統實作 22
4.4.1 Platform 22
4.4.2 Journaling Component 22 Logger 23 Update Daemon 24
4.4.3 VMM Component 25 Update Manager 25 Protection Manager 27
4.4.4 Recovery Implementation 28
第五章 效能評估 30
5.1 實驗環境 30
5.2 Runtime Performance of Single Guest Domain 30
5.2.1 Micro-benchmark: Bonnie++ 30
5.2.2 Un-tar/Make Linux Kernel 33
5.2.3 Macro-benchmark: Postmark 34
5.3 Runtime Performance of Multiple Guest Domains 35
5.4 Filesystem Recovery Performance 37
5.5 CPU/Memory Overhead Measurement 38
第六章 結論及未來展望 41
6.1 Conclusions 41
6.2 Future Works 41
參考文獻 42
[1]Advanced Micro Devices Inc., Pacifica Virtualization Technology, available at http://developer.amd.com/assets/WinHEC2005_Pacifica_Virtualization.pdf, 2005
[2]M. Baker, S. Asami, E. Deprit, J. Ousterhout and M. Seltzer, “Non-volatile Memory for Fast, Reliable Filesystems,” In Proceedings of the 5-th International Conference on Architectural Support for Programming Languages and Operating Systems, pp. 10-22, Oct. 1992.
[3]P. Barham, B. Dragovic, K. Fraser, S. Hand, T. Harris, A. Ho, R. Neugebauery, I. Pratt and A. Warfeld, “Xen and the Art of Virtualization,” In Proceedings of the ACM Symposium on Operating Systems Principles, pp. 164-177, Oct. 2003.
[4]T. Bressoud and F. Schneider, “Hypervisor-based fault-tolerance,” ACM Transactions on Computer Systems, Vol. 14, No. 1, pp. 80-107, Feb. 1996.
[5]R. Card, T. Ts'o and S. Tweedie, “Design and Implementation of the Second Extended Filesystem.” In Proceedings of the 1st Dutch International Symposium on Linux, 1994.
[6]P. Chen, W. Ng, S. Chandra, C. Aycock, G. Rajamani and D. Lowell, “The Rio File Cache: Surviving Operating System Crashes,” In Proceedings of the 7th International Conference on Architectural Support for Programming Languages and Operating Systems, pp. 74-83, 1996.
[7]L. Cherkasova and R. Gardner, “Measuring CPU Overhead for I/O Processing in the Xen Virtual Machine Monitor," In Proceedings of the USENIX Annual Technical Conference, pp. 387-390, Apr. 2005.
[8]S. Chutani, O. Anderson, M. Kazer, B. Leverett, A. Mason and R. Sidebotham, “The Episode File System,” In Proceedings of the Winter USENIX Technical Conference, pp. 43-60, Jan. 1992.
[9]R. Coker, The bonnie++ benchmark, available at http://www.coker.com.au/bonnie++, 2001
[10]H. Custer, “Inside the Windows NT Filesystem,” Microsoft Press. ISBN 155615660X, 1994.
[11]J. Dike, “A User-mode Port of the Linux kernel,” In the Proceedings of the Annual Linux Showcase, pp. 63-72, Oct. 2000.
[12]R. Duncan, “Design Goals and Implementation of the New High Performance File System,” Microsoft Systems Journal, Vol. 4, No. 5, pp. 1-13, 1989.
[13]K. Elkhardt and R. Bayer, “A Database Cache for High Performance and Fast Restart in Database Systems,” ACM Transactions on Database Systems, Vol. 9, No. 4, pp. 503-525, Dec. 1984.
[14]K. Fraser, S. Hand, R. Neugebauer, I. Pratt, A. Wareld and M. Williamson, “Reconstructing I/O," Technical Report UCAM-CL-TR-596, Cambridge University, Aug. 2004.
[15]K. Fraser, S. Hand, R. Neugebauer, I. Pratt, A. Wareld and M. Williamson, “Safe Hardware Access with the Xen Virtual Machine Monitor," In Proceedings of the 1st Workshop on Operating System and Architectural Support for the On-Demand IT Infrastructure, Oct. 2004.
[16]R. Goldberg, “Survey of Virtual Machine Research,” IEEE Computer Magazine, Vol. 7 No.6, pp. 34-45, Jun. 1974.
[17]J. Gray and A. Reuter, Transaction Processing: Concepts and Techniques. San Mateo, CA: Morgan Kaufmann, 1993.
[18]P. Gum, “System/370 Extended Architecture: Facilities for Virtual Machines,” IBM Journal of Research and Development, Vol. 27, No. 6, pp. 530-544, Nov. 1983.
[19]Intel Inc., Vanderpool Virtualization Technology, available at http://www.intel.com/technology/computing/vptech/, 2005
[20]International Business Machines Inc., JFS, available at http://jfs.sourceforge.net/, 1990
[21]X. Jiang and D. Xu, “Collapsar: A VM-based Architecture for Network Attack Detention Center,” In Proceedings of the USENIX Security Symposium, pp. 15-28, Aug. 2004.
[22]J. Katcher, “Postmark: A New File System Benchmark,” Technical Report TR3022, Network Appliance Inc., Oct. 1997.
[23]K. Kourai and S. Chiba, “HyperSpector: Virtual Distributed Monitoring Environments for Secure Intrusion Detection,” In Proceedings of the 1st ACM/USENIX international conference on Virtual execution environments, pp. 197-207, Jun. 2005.
[24]M. McKusick and G. Ganger, “Soft Updates: A Technique for Eliminating Most Synchronous Writes in the Fast Filesystem.” In Proceedings of the 1999 USENIX Annual Technical Conference, FREENIX Track, pp. 1-18, 1999.
[25]M. McKusick, W. Joy, S. Leffler and R. Fabry, “A Fast Filesystem for UNIX,” ACM Transactions on Computer Systems, Vol. 2, No.3, pp. 181-197, Aug. 1984.
[26]A. Menon, J. Santos, Y. Turner, G. Janakiraman and W. Zwenepoel, “Diagnosing Performance Overheads in the Xen Virtual,” In Proceedings of the 1st ACM/USENIX International Conference on Virtual Execution Environments, pp. 13-23, Jun. 2005.
[27]Microsoft Inc., Virtual PC, available at
http://www.microsoft.com/windows/virtualpc/default.mspx, 1997
[28]D. Patterson, P. Chen, G. Gibson and R. Katz, “Introduction to Redundant Arrays of Inexpensive Disks (RAID),” In Digest of Papers for 34th IEEE Computer Society International Conference, pp. 112 -117, 1989.
[29]V. Prabhakaran, A. Arpaci-Dusseau and R. Arpaci-Dusseau, “Analysis and Evolution of Journaling File Systems,” In Proceedings of the USENIX Annual Technical Conference Anaheim, pp. 105-120, Apr. 2005.
[30]H. Reiser, ReiserFS, available at http://www.namesys.com/, 2001
[31]J. Robin and C. Irvine, “Analysis of the Intel Pentium’s Ability to Support a Secure Virtual Machine Monitor,” In the Proceedings of 9th USENIX Security Symposium, pp. 129-144, Aug. 2000.
[32]R. Rose, “Survey of System Virtualization Techniques,” available at http://www.robertwrose.com/vita/rose-virtualization.pdf, 2004.
[33]M. Rosenblum and T. Garfinkel, “Virtual Machine Monitors: Current Technology and Future Trends,” IEEE Computer Society, Vol 38, No. 5, pp. 39-47, May 2005.
[34]M. Rosenblum and J. Ousterhout, “The Design and Implementation of a Log-Structured Filesystem,” In Proceedings of the symposium on Operating System Principles, pp. 1-15, Oct. 1991.
[35]R. Shiveley, “Consolidation Strategies for Intel Processor-based Servers,” Intel white paper, Feb. 2006
[36]Silicon Graphics Inc., XFS, available at http://oss.sgi.com/projects/xfs/, 1994
[37]L. Soares, O. Krieger and D. Silva, “Meta-data Snapshotting: A Simple Mechanism for Filesystem Consistency," International Workshop on Storage Network Architecture and Parallel I/Os, pp. 1-12, Sep. 2003.
[38]J. Sugerman, G. Venkitachalam and B. Lim, “Virtualizing I/O Devices on VMware Workstation’s Hosted Virtual Machine Monitor,” In Proceedings of the USENIX Annual Technical Conference, pp. 1-14, Jun. 2001.
[39]T. Ts'o and S. Tweedie, “Planned Extensions to the Linux Ext2/Ext3 Filesystem,” In Proceedings of the USENIX Annual Technical Conference, FREENIX Track, pp. 234-244, 2002.
[40]C. Waldspurger, “Memory Resource Management in VMware ESX Server,” In Proceedings of the 5th Symposium on Operating Systems Design and Implementation, pp. 181-194, Dec. 2002.
[41]A. Whitaker, R. Cox and S. Gribble, “Configuration Debugging as Search: Finding the Needle in the Haystack,” In Proceedings of the Symposium on Operating Systems Design and Implementation, pp. 77-90, Dec. 2004.
[42]A. Whitaker, M. Shaw and S. Gribble, “Scale and Performance in the Denali Isolation Kernel,” In Proceedings of 5th Symposium on Operating System Design and Implementation, pp. 195-209, Dec. 2002.
[43]M. Williamson, “Extreme Paravirtualisation: Beyond arch/xen,” available at http://www.cambridge.intel-research.net/~mwilli2/proposal_final.pdf, Jul. 2005
[44]M. Williamson, “1st Year Progress Report,” available at http://www.cambridge.intel-research.net/~mwilli2/report_final.pdf, Jul. 2005
第一頁 上一頁 下一頁 最後一頁 top