SONMAS: 多維屬性空間適用之結構化套疊網路｜國立清華大學博碩士論文庫

簡易檢索 / 詳目顯示

回結果列表

研究生：	陳秀琴 Hsiu-Chin Chen
論文名稱：	SONMAS: 多維屬性空間適用之結構化套疊網路 SONMAS: A Structured Overlay Network for Multidimensional Attribute Space
指導教授：	金仲達 Chung-Ta King
口試委員:
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 資訊系統與應用研究所 Institute of Information Systems and Applications
論文出版年：	2006
畢業學年度：	94
語文別：	英文
論文頁數：	59
中文關鍵詞：	結構化套疊網路、屬性空間、多重屬性、點對點、範圍搜尋、近似值搜尋
外文關鍵詞：	peer-to-peer, structured overlay network, attribute based, multi-dimension, SONMAS, attribute space, multiple attributes, attribute collision, attribute-ID-hybrid key, attribute-ID-hybrid space, range search, proximal value search, interest space, multiple interests
相關次數：	點閱：2 下載：0
分享至:	分享至facebook 分享至twitter

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

本論文中提出了一個適用於多維屬性空間的結構化的點對點套疊網路― SONMAS (Structured Overlay Network for Multidimensional Attribute Space)。在SONMAS的應用場景中，所有的節點(node)均具有一組可量化的多維屬性，並依此屬性建立彼此的連結關係; 透過SONMAS所提供的搜尋功能，我們可以有效地找到滿足某特定屬性範圍的一群節點，以供後續的運算之用。傳統的點對點系統不外乎採用類似Gnutella的連結，以flooding的手法作搜尋，或是在一結構化的連結之上建構一個分散式的雜湊表(DHT)以達成搜尋的目的。前者面臨的問題是過量的flooding訊息流量，而後者的問題是無法有效地支援範圍搜尋或連續值搜尋。由於SONMAS是直接依據節點之屬性而建立結構化連結，因此可以自然地避免掉傳統的點對點系統所遭遇到的問題。對於屬性空間中可能發生的屬性衝撞問題，SONMAS利用一個屬性-識別號混成鍵的概念，將每一個節點從屬性空間映射到屬性-識別號混成空間中，成為一個個獨特的點。SONMAS將屬性-識別號混成空間切割成格子與層層的子格，並依據節點對格子的從屬關係而建立結構化的連結。在低維度屬性空間中SONMAS有良好的表現，模擬的結果顯示SONMAS的繞徑效能與維護訊息量成本對於網路大小具有可縮放性(scalability)。同時，模擬的結果也顯示另人滿意的容錯性、復原力及可靠性。

This thesis introduces SONMAS (Structured Overlay Network for Multidimensional Attribute Space), a multidimensional attribute-based peer-to-peer (P2P) system. SONMAS considers applications in which peers characterized by multidimensional attributes are discovered to satisfy given range queries. Traditional P2P systems either adopt a Gnutella-like interconnection and use flooding to search for matching peers, or implement a distributed hash table (DHT) on top of a structuralized interconnection. The former suffers from excessive flooding traffic, while the latter has difficulty of supporting range queries. SONMAS avoids these problems by interconnecting the peers according to their attributes. To deal with possible attribute collisions, SONMAS uses a special concept of attribute-ID-hybrid space to map each node to a unique point in the multidimensional space. SONMAS divides the attribute-ID-hybrid space into different levels of cells and interconnect the nodes according to their memberships to the cells. SONMAS is expected to work for applications with low dimensional attributes. Simulation results confirm its scalable routing efficiency and maintenance overhead as well as good fault-tolerance and resilience.

Abstract ................................................ i
Table of Contents ...................................... ii
List of Figures ........................................ iv
Chapter 1 Introduction .................................. 1
1 Motivation ........................................ 2
2 The Research ...................................... 4
3 Thesis Organization ............................... 5
Chapter 2 Related Works ................................. 6
1 DHTs ...............................................6
1.1 CAN ........................................... 7
1.2 Chord ......................................... 8
1.3 Pastry ........................................ 8
2 Murk .............................................. 9
3 Skip Graph ....................................... 10
4 Skip Quad-tree ................................... 11
Chapter 3 SONMAS system design ......................... 13
1 Attribute-ID-hybrid Space ........................ 13
2 Space Division and Interconnection rules ......... 15
2.1 Division of Attribute-ID-hybrid Space ........ 15
2.2 Interconnection Rules ........................ 18
3 Basic Operations ................................. 20
3.1 Routing ...................................... 20
3.2 Join/Quit .................................... 22
3.3 Attribute Shift .............................. 25
3.4 Query ........................................ 26
4 Maintenance ...................................... 29
4.1 Probing ...................................... 30
4.2 Routing Table Exchange ....................... 31
4.3 Backups ...................................... 33
4.4 Proximity Neighbor Selection and Hot Spot
          Avoidance..................................... 34
Chapter 4 Simulation ............................................. 36
1 The Simulator .................................... 36
1.1 Core Engine .................................. 37
1.2 Application-level Protocol ................... 37
2 Experiments ...................................... 38
2.1 Time Efficiency and Traffic Overhead ......... 39
2.1.1 Method .................................... 40
2.1.2 Results ................................... 40
2.2 Reliability under Dynamic Environments ....... 43
2.2.1 Method .................................... 44
2.2.2 Results ................................... 45
2.3 Query ........................................ 49
2.3.1 Method .................................... 49
2.3.2 Results ................................... 50
3 Summary .......................................... 52
Chapter 5 Discussions .................................. 54
1 The Limitations of SONMAS ........................ 54
2 Hash-Based or Attribute-Based .................... 55
Chapter 6 Conclusions .................................. 57
Bibliography ........................................... 58

                                

[1] Ittai Abraham, James Aspnes, and Jian Yuan, “Skip B-
Trees”, Proc. Ninth International Conference on
Principals of Distributed Systems, pp. 284-295, Dec.
2005.
[2] Lars Arge, David Eppstein, and Michael T. Goodrich,
“SkipWebs: Efficient Distributed Data Structures for
MultiDimensional Data Sets”, Proc. of the Twenty-
fourth Annual ACM SIGACT-SIGOPS Symposium on Principles
of Distributed Computing (PODC), 2005.
[3] James Aspnes and Gauri Shah, “Skip Graphs”, Proc.
Fourteenth Annual ACM-SIAM Symposium on Discrete
Algorithms, pp. 384-393, Jan. 2002
[4] I. Clarke, O. Sandberg, B. Wiley, and T. W. Hong,
“Freenet: A Distributed Anonymous Information Storage
and Retrieval System”, Proc. Workshop on Design Issues
in Anonymity and Unobservability, pages 311–320, July
2000. ICSI, Berkeley, CA, USA.
[5] D. Eppstein, M. T. Goodrich, and J. Z. Sun, “The Skip
Quadtree: A Simple Dynamic Data Structure for
Multidimensional Data”, Proc. 21st ACM Symp. On
Computational Geometry (SCG), 2005
[6] Prasanna Ganesan, Beverly Yang, and Hector
GarciaMolina, “One Torus to Rule Them All:
Multidimensional Queries in P2P Systems”, Proc. of the
Seventh International Workshop on the Web and Databases
(WebDB), 2004.
[7] Nicholas J. A. Harvey, Michael B. Jones, Stefan Saroiu,
Marvin Theimer, and Alec Wolman, “SkipNet: A Scalable
Overlay Network with Practical Locality Properties”,
Proc. of the Fourth USENIX Symposium on Internet
Technologies and Systems (USITS), pages 113–126,
Seattle, WA, USA, March 2003.
[8] Sylvia Ratnasamy, Paul Francis, Mark Handley, Richard
Karp, and Scott Shenker, “A Scalable Content-
Addressable Network”, Proc. ACM Symposium on
Communications Architectures and Protocols (SIGCOMM),
pages 161–172, San Diego, CA, USA, August 2001.
[9] Sean Rhea, Dennis Geels, Timothy Roscoe, and John
Kubiatowicz, “Handling Churn in a DHT”, Proc. 2004
USENIX Technical Conference, 2004.
[10] A. Rowstron and P. Druschel, “Pastry: Scalable,
Decentralized Object Location and Routing for Large-
scale Peer-to-peer Systems”, Lecture Notes in
Computer Science, vol.2218, pp. 161-172, November 2001.
[11] I. Stoica, R. Morris, D. Karger, M. F. Kaashoek, and
H. Balakrishnan, “Chord: A Scalable Peer-to-Peer
Lookup Service for Internet Applications”, Proc. of
the International Conference on Applications,
Technologies, Architectures, and Protocols for
Computer Communications. ACM press, August 2001, pp.
149-160.
[12] Napster. http://www.napster.com/
[13] Gnutella. http://www.gnutella.com/

全文公開日期本全文未授權公開 (校內網路)
全文公開日期本全文未授權公開 (校外網路)

簡易檢索 / 詳目顯示

相關論文