(3.238.130.97) 您好!臺灣時間:2021/05/09 03:34
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:::

我願授權國圖
: 
twitterline
研究生:董星甫
研究生(外文):Hsing-Fu Tung
論文名稱:高效能分散式檔案系統的設計實作與效能量測
論文名稱(外文):Design, Implementation, and Performance Evaluation of a High Performance Distributed File System
指導教授:雷欽隆雷欽隆引用關係
指導教授(外文):Chin-Laung Lei
學位類別:碩士
校院名稱:國立臺灣大學
系所名稱:電機工程學研究所
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:2004
畢業學年度:92
語文別:英文
論文頁數:77
中文關鍵詞:效能分散式檔案系統
外文關鍵詞:Distributed file systemperformance
相關次數:
  • 被引用被引用:0
  • 點閱點閱:231
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
分散式檔案系統是將檔案集中儲存在伺服器,讓客戶端透過網路存取伺服器
上的檔案,藉此可提供更有效率的檔案使用。很多因素會影響分散式檔案系統的
效能,其中一個就是客戶端與伺服端之間的網路傳輸。
隨著各項軟硬體技術的成熟,網際網路傳輸的速度與品質相較於以往有長足
的進步。在此同時,為了有效運用充足的網路資源,各種高效能的通訊協定陸續
被提出,這一類的協定具有的共通點在於以高頻寬、高網路延遲的環境為目標,
致力於增加頻寬使用效率,並且在網路延遲增加時,仍能維持穩定的效能。相對
的,原本並非針對這種環境設計的分散式檔案系統在高頻寬、高網路延遲的環境
下,一方面對於頻寬的利用不盡理想,同時效能更會隨著網路延遲的增加而有顯
著的下滑,進而影響整個檔案系統的表現。
綜合以上各項因素,這篇論文提出一個新的分散式檔案系統,將高效能通訊
協定整合進入分散式檔案系統,針對檔案的網路傳輸,克服原本在高頻寬、高網
路延遲環境下的缺陷。並且經由實作與測試,證明經過整合的分散式作業系統確
實在檔案的網路傳輸上表現出顯著的進步。
分散式檔案系統是將檔案集中儲存在伺服器,讓客戶端透過網路存取伺服器
上的檔案,藉此可提供更有效率的檔案使用。很多因素會影響分散式檔案系統的
效能,其中一個就是客戶端與伺服端之間的網路傳輸。
隨著各項軟硬體技術的成熟,網際網路傳輸的速度與品質相較於以往有長足
的進步。在此同時,為了有效運用充足的網路資源,各種高效能的通訊協定陸續
被提出,這一類的協定具有的共通點在於以高頻寬、高網路延遲的環境為目標,
致力於增加頻寬使用效率,並且在網路延遲增加時,仍能維持穩定的效能。相對
的,原本並非針對這種環境設計的分散式檔案系統在高頻寬、高網路延遲的環境
下,一方面對於頻寬的利用不盡理想,同時效能更會隨著網路延遲的增加而有顯
著的下滑,進而影響整個檔案系統的表現。
綜合以上各項因素,這篇論文提出一個新的分散式檔案系統,將高效能通訊
協定整合進入分散式檔案系統,針對檔案的網路傳輸,克服原本在高頻寬、高網
路延遲環境下的缺陷。並且經由實作與測試,證明經過整合的分散式作業系統確
實在檔案的網路傳輸上表現出顯著的進步。
1 Introduction 1
2 Related Works 7
2.1 Distributed File Systems: Case Studies . . . . . . . . . . . . . 7
2.1.1 The Sun network ‾le system (NFS) . . . . . . . . . . . 7
2.1.2 The Sprite File System (SFS) . . . . . . . . . . . . . . 9
2.2 High Performance Transport Protocols . . . . . . . . . . . . . 10
3 Preliminaries 15
3.1 The Coda File System . . . . . . . . . . . . . . . . . . . . . . 15
3.1.1 Client and Server Architecture . . . . . . . . . . . . . . 16
3.1.2 Naming, Volumes, and Server Replication . . . . . . . 18
3.1.3 Caching and Disconnected Operation . . . . . . . . . . 20
3.1.4 The Remote Procedure Call and RPC2 . . . . . . . . . 23
3.1.5 SFTP Protocol . . . . . . . . . . . . . . . . . . . . . . 27
3.2 UDT/SABUL Protocol . . . . . . . . . . . . . . . . . . . . . . 34
3.2.1 Protocol Overview/System Architecture . . . . . . . . 34
3.2.2 Congestion Control . . . . . . . . . . . . . . . . . . . . 40
3.2.3 Communication Algorithm . . . . . . . . . . . . . . . . 44
4 Analysis and Implementation 51
4.1 Analysis of SFTP and UDT . . . . . . . . . . . . . . . . . . . 51
4.2 Design and Implementation of Coda-UDT . . . . . . . . . . . 54
5 Experiments 59
5.1 Experiment Environment . . . . . . . . . . . . . . . . . . . . . 60
5.2 Experiment Results . . . . . . . . . . . . . . . . . . . . . . . . 61
5.2.1 Experiments on Bandwidth and Latency . . . . . . . . 62
5.2.2 Experiments on Packet Loss . . . . . . . . . . . . . . . 65
6 Conclusions 67
[1] IBM BladeCenter HS20. http://www5.pc.ibm.com/us/products.
nsf/$wwwPartNumLookup/ 867841X.
[2] NISTNET. http://snad.ncsl.nist.gov/itg/nistnet/.
[3] Tsunami. http://www.anml.iu.edu/anmlresearch.html, DEC 2002.
[4] Bill Allcock, Joe Bester, John Bresnahan, Ann L. Chervenak, Ian Fos-
ter, Carl Kesselman, Sam Meder, Veronika Nefedova, Darcy Ques-
nal, and Steven Tuecke. Data management and transfer in high-
performance computational grid environments. Parallel Computing
Journal, 28(5):749{771, May 2002.
[5] Andrew D. Birrell and Bruce Jay Nelson. Implementing remote proce-
dure calls. ACM Transaction on Computer System, 2(1):39{59, 1984.
[6] John Bloomer. Power programming with RPC. O''Reilly & Associates,
Inc., 1992.
[7] Peter J. Braam. The coda distributed ‾le system. Linux Journal,
1998(50es):6, 1998.
[8] Peter J. Braam and RobertWatson. RPC2 A snapshot of understanding.
http://www.coda.cs.cmu.edu/doc/ppt/rpc2.ppt.
[9] L. Brakmo and L. Peterson. End-to-end congestion avoidance on a
global Internet. IEEE Journal on Selected Areas in Communication,
13(8):1465{1480, 1995.
[10] Mark Carson and Darrin Santay. NIST Net: a Linux-based network
emulation tool. SIGCOMM Comput. Commun. Rev., 33(3):111{126,
2003.
[11] David Cheriton. Vmtp: Versatile message transaction protocol. RFC
1045, FEB 1998. ftp://ftp.rfc-editor.org/in-notes/rfc1045.
txt.
[12] Randy Chow. Distributed Operating Systems & Algorithms, chapter 6.
Addison-Wesley, Inc., 1997.
[13] D. D. Clark, M. L. Lambert, and L. Zhang. NETBLT: a high throughput
transport protocol. SIGCOMM Comput. Commun. Rev., 17(5):353{359,
1987.
[14] T. H. Dineen, P. J. Leach, N. W. Mishkin, J. N. Pato, and G. L.
Wyant. The network computing architecture and system: An envi-
ronment for developing distributed applications. In Proceedings of the
Summer USENIX Conference, pages 385{398. USENIX, 1987.
[15] Aaron Falk, Ted Faber, Joseph Bannister, Andrew Chien, Robert Gross-
man, and Jason Leigh. Transport protocols for high performance. Com-
munications of the ACM, 46(11):42{49, 2003.
[16] S. Floyd, J. Mahdavi, M. Mathis, and M. Podolsky. An extension to the
selective acknowledgement (SACK) option for TCP. RFC 2883, JUL
2000. ftp://ftp.rfc-editor.org/in-notes/rfc2883.txt.
[17] Sally Floyd. HighSpeed TCP for large congestion windows.
IETF Internet Draft, 2003. http://www.icir.org/floyd/papers/
draft-floyd-tcp-highspeed-02.txt.
[18] M. Gerla, M. Y. Sanadidi, R. Wang, A. Zanella, C. Casetti, and S. Mas-
colo. TCP Westwood: congestion window control using bandwidth es-
timation. In Proceedings of the IEEE Global Telecommunications Con-
ference, volume 3, pages 1698{1702, 2001.
[19] Yuhong Gu, Xinwei Hong, and Marco Mazzucco. Sabul: A high perfor-
mance data transfer protocol. Submitted to IEEE COMMUNICATION
LETTERS, 2003.
[20] Yunhong Gu, Robert Grossman, Xinwei Hong, and Marco Mazzucco.
Using UDP for reliable data transfer over high bandwidth-delay product
networks. Submitted to Computer Communication Review, 2003.
[21] Yunhong Gu and Robert L. Grossman. UDT: A transport protocol for
data intensive applications. IETF Internet Draft, 2003.
[22] M. Mazzucco Y. Pan Q. Zhang H. Sivakumar, R. L. Grossman. Simple
available bandwidth utilization library for high-speed wide area net-
works. Journal of Supercomputing, to appear, but not found, 2001.
[23] V. Jacobson, R. Braden, D. Borman, and Cray Research. TCP exten-
sions for high performance. RFC 1323, May 1992.
[24] Amit Jain and Sally Floyd. Quick-Start for TCP and IP.
IETF Internet Draft, 2002. http://www.icir.org/floyd/papers/
draft-amit-quick-start-02.txt.
[25] Dina Katabi, Mark Handley, and Charlie Rohrs. Congestion control
for high bandwidth-delay product networks. In Proceedings of the 2002
conference on Applications, technologies, architectures, and protocols for
computer communications, pages 89{102. ACM Press, 2002.
[26] Tom Kelly. Scalable TCP, 2002. http://www-lce.eng.cam.ac.uk/
»ctk21/scalable/.
[27] Tom Kelly. Scalable TCP: Improving performance in highspeed wide
area networks. http://www-lce.eng.cam.ac.uk/»ctk21/papers/
scalable improve hswan.pdf, 2002.
[28] James J. Kistler and M. Satyanarayanan. Disconnected operation in the
Coda ‾le system. ACM Transaction on Computer System, 10(1):3{25,
1992.
[29] M. Kong, T. H. Dineen, P. J. Leach, E. A. Martin, N. W. Mishkin, J. N.
Pato, and G. L. Wyant. Network Computing System Reference Manual.
Prentice Hall, 1990.
[30] J. Leigh, O. Yu, D. Schonfeld, R. Ansari, E. He, A. Nayak, J. Ge, N. Kr-
ishnaprasad, K. Park, Y. Cho, L. Hu, R. Fang, A. Verlo, L. Winkler,
and T. A. DeFanti. Adaptive networking for tele-immersion. In Pro-
ceedings of the 5th Immersive Projection Technology/7th Eurographics
Virtual Environments Conference, pages 199{208, May 2001.
[31] Mark K. Lottor. Simple ‾le transfer protocol. RFC 913, 1984.
[32] L. B. Mummert, M. R. Ebling, and M. Satyanarayanan. Exploiting
weak connectivity for mobile ‾le access. In Proceedings of the ‾fteenth
ACM symposium on Operating systems principles, pages 143{155. ACM
Press, 1995.
[33] J.K. Ousterhout, A.R. Cherenson, F. Douglis, M.N. Nelson, and B.B.
Welch. The Sprite network operating system. IEEE Computer, 21(2):23{
36, FEB 1988.
[34] F. Paganini, Z.Wang, S. H. Low, and J. C. Doyle. A new TCP/AQM for
stable operation in fast networks. In Proceedings of the IEEE Twenty-
Second Annual Joint Conference of the IEEE Computer and Communi-
cations Societies, volume 1, pages 96{105, APR 2003.
[35] K. Ramakrishnan, S. Floyd, and D. Black. The addition of explicit
congestion noti‾cation (ECN) to IP. RFC 3168, SEP 2001. ftp://ftp.
rfc-editor.org/in-notes/rfc3168.txt.
[36] Russel Sandberg, David Goldberg, Steve Kleiman, Dan Walsh, and Bob
Lyon. Design and implementation of the Sun network ‾lesystem. In Pro-
ceedings of the Summer USENIX Conference, pages 119{130. USENIX,
1985.
[37] M. Satyanarayanan, Richard Draves, James Kistler, Anders Klemets,
Qi Lu, Lily Mummert, David Nichols, Larry Raper, Gowthami Rajen-
dran, Jonathan Rosenberg, and Ellen H. Siegel. RPC2 User Guide and
Reference Manual.
[38] M. Satyanarayanan, Maria R. Ebling, Peter J. Braam, and Jan Harkes.
Coda File System User and System Administrators Manual, 1997.
[39] M. Satyanarayanan, J.J. Kistler, P. Kumar, M.E. Okasaki, E.H. Siegel,
and D.C. Steere. Coda: a highly available ‾le system for a distributed
workstation environment. IEEE Transactions on Computers, 39:447{
459, 1990.
[40] M. Satyanarayanan and Ellen H. Siegel. Parallel communication in
a large distributed environment. IEEE Transactions on Computers,
39(3):328{348, 1990.
[41] Mukesh Singhal, Niranjan G. Shivaratri, and Niranjan Shivaratro. Ad-
vanced Concepts in Operating Systems, chapter 9. McGraw-Hill, Inc.,
1994.
[42] H. Sivakumar, S. Bailey, and R. L. Grossman. PSockets: the case for
application-level network striping for data intensive applications using
high speed wide area networks. In Proceedings of the 2000 ACM/IEEE
conference on Supercomputing (CDROM), page 37. IEEE Computer So-
ciety, 2000.
[43] W. Richard Stevens. UNIX network programming, chapter 18. Prentice-
Hall, Inc., 1990.
[44] Inc. Sun Microsystems. RPC: Remote procedure call protocol speci‾ca-
tion version 2. RFC 1057, 1987.
[45] Inc. Sun Microsystems. XDR: External data representation standard.
RFC 1014, 1987.
[46] Andrew S. Tanenbaum. Distributed Operating Systems, chapter 5.
Prentice-Hall, Inc., 1995.
[47] Brent Welch and John K. Ousterhout. Pre‾x tables: A simple mech-
anism for locating ‾les in a distributed system. In Proceedings of the
6th International Conference on Distributed Computing Systems, pages
184{189. IEEE Computer Society Press, May 1986.
BIBLIOGRAPHY 77
[48] Xerox. Courier: The remote procedure call protocol. XSIS 038112,
Xerox Corp., 1981.
[49] L. Zahn, T. H. Dineen, P. J. Leach, E. A. Martin, N. W. Mishkin, J. N.
Pato, and G. L. Wyant. Network Computing Architecture. Prentice Hall,
1990.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔