本論文提出全球資訊網效能的改善方法：使用合作式快取技術尋找網頁物件所在位置減少網路的資料傳輸，改善全球資訊網的效能。
雜湊繞徑法〈Hash Routing〉將網頁物件的URL 經過雜湊函式得到唯一數值，依據此數值選取編號相同的快取伺服器。優點是減少網頁物件的重複儲存，提高快取命中率；缺點則是當快取伺服器當機無法運作時，會發生嚴重問題〈Single-Point-Of-Failure and Cache Disruption〉。目錄摘要法〈Directory-based Digest〉將網頁物件所存放的快取伺服器之編號，記載於目錄摘要，以便查表得知網頁物件的位置。優點是可以縮短查詢時間；缺點則是目錄摘要法所使用的記憶容量與儲存網頁物件的數量成正比，將會加重網路與快取伺服器的負載。
我們將以上兩者合併使用，取其利而避其害。故記載於目錄摘要中的網頁物件必須謹慎選取，不宜過多。本論文依據網頁物件的價值高低，分為兩類，價值高者，適用目錄摘要法；價值低者，適用雜湊繞徑法。以上兩種方法合併使用，在提高快取命中率與降低網路負載間尋求平衡，並改善全球資訊網的效能。
經過實驗模擬的驗證，並與其他傳統方法比較。採用合作式快取技術，利用雜湊繞徑及目錄摘要的混合方法可在提高快取命中率與網路負載間取得平衡。

Nowadays, Web caching technique has been widely used and become one of the most promising ways to reduce network traffic, server load, and user-experienced latency while users surf the Web. In the context of traditional systems, caching techniques have been extensively studied. However, these techniques are not directly applicable to Web due to larger size of working set and cache storage in proxies. Many studies have presented their approaches to improving the performance of Web caching. Two of most representative approaches are hash routing [25] and directory-base digest [12]. Hash routing provides amapping fromthe URL of object to the location of proxy, which has the cached object, while directory-based digest records the pairs of proxy locations and object URLs for answering the query when local misses occur in any proxy. Hash routing can best utilize storage space by eliminating duplicated objects among proxies,while directory-based digest allows object replicas among proxies to resist proxy failures. These two conventional approaches have complementary tradeoffs.
In this thesis, a comprehensive approach to cooperative caching for Web proxies, using a combination of hash routing and directory-based digest, is presented. Our approach tends to subsume these widely used approaches and thus gives a spectrum of trade-off between the overall hit ratio and its associated overhead. Through the simulations using real-life proxy traces, the performance and overhead of our proposed mechanism were evaluated. The experimental results showed that our approach outperforms the previous competitors.

Abstract iii
中文摘要 v
Definition vii
Table of Contents ix
List of Tables xii
List of Figures xiii
1 Introduction 1
1.1 Research Background and Motivation . . . . . . . . . . . . . . . . . 3
1.2 Our Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Organization of this Thesis . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Preliminary 7
2.1 Workload Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.1 Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1.2 Workload Traces . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1.3 Object Access Popularity: Zipf-like Distribution . . . . . . . 9
2.1.4 Temporal Locality . . . . . . . . . . . . . . . . . . . . . . . . 10
2.2 Performance Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.2.1 Resilient to Proxy Failures . . . . . . . . . . . . . . . . . . . . 11
2.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3 Cooperative Web Caching 13
3.1 Object Locating and Request Redirection . . . . . . . . . . . . . . . 14
3.1.1 Direct or Indirect Query . . . . . . . . . . . . . . . . . . . . . 14
3.2 Object Retrieval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2.1 Relayed or Non-relayed Retrieval . . . . . . . . . . . . . . . 15
3.3 Object Placement and Replication . . . . . . . . . . . . . . . . . . . . 16
4 Related Work 17
4.1 ICP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.2 Hash Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
4.3 Latency-Sensitive Hash routing (LSH) . . . . . . . . . . . . . . . . . 18
4.4 Directory-based Digest . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.5 CRISP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.6 Hash-based Query Caching Method (HQCM) . . . . . . . . . . . . . 20
4.7 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
5 Hybrid Caching: A Comprehensive Approach 22
5.1 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.2 Solutions and Design Principle . . . . . . . . . . . . . . . . . . . . . 25
5.2.1 Directory-based Digest . . . . . . . . . . . . . . . . . . . . . . 26
5.2.2 Valuable Objects . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.2.3 Hash Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
6 Simulation Evaluations 31
6.1 Experimental Methodology . . . . . . . . . . . . . . . . . . . . . . . 31
6.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
6.3 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
7 Conclusions and Future Work 38
References 40

[1] Almeida, V., A. Bestavros, M. Crovella, and A. de Oliveira, “Characterizing reference locality in theWeb,” in Proceedings of IEEE Conf. Parallel and Distributed Information Systems (PDIS'96), Pages 92–107, December 1996.
[2] Asaka, T., H. Miwa, and Y. Tanaka, “Distributed Web caching using hash-based query caching method,” in Proceedings of Int'l Conf. Control Applications, August 1999.
[3] Asaka, T., H. Miwa, and Y. Tanaka, “Hash-based query caching method for distributed Web caching in wide area networks,” IEICE Trans. Commun., Volume E82-B, Number 6, Pages 907–914, June 1999.
[4] Baeza-Yates, R., C. Castillo, and V. Lopez, “Characteristics of the web of spain,” Cybermetrics, Volume 9, Number 1, Pages 138–141, 2005.
[5] Bahn, H., K. Koh, S. H. Noh, and S. L. Min, “Efficient replacement for non-uniform objects in Web caches,” IEEE Computer, Volume 35, Number 6, Pages 68–73, June 2002.
[6] Berners-Lee, T., R. Cailliau, J.-F. Groff, and B. Pollermann, “World-Wide Web: The information universe,” Electronic Networking: Research, Applications, and Policy, Volume 2, Number 1, Pages 52–58, Spring 1992. [Online]. Available: http://www.w3.org/History/1992/ENRAP/Article_9202.ps.
[7] Breslau, L., P. Cao, L. Fan, G. Philips, and S. Shenker, “Web caching and Zipf-like distributions: Evidence and implications,” in Proceedings of IEEE INFOCOM'99, Pages 126–134, March 1999.
[8] Busari, M. and C. Williamson, “ProWGen: A synthetic workload generation tool for the simulation evaluation ofWeb proxy caches,” Computer Networks, Volume 38, Number 6, Pages 779–794, April 2002.
[9] Cano, J.-C., A. Pont, J. Sahuquillo, and J.-A. Gil, “The differences between distributed shared memory caching and proxy caching,” IEEE Concurrency, Volume 8, Number 3, Pages 45–47, July-September 2000.
[10] Chankhunthod, A., P. B. Danzig, C. Neerdaels, M. J. Schwartz, and K. J.Worrell, “A hierarchical Internet object cache,” Technical Report 95-611, Computer Science Department, University of Southern California, Los Angels, CA,March 1995. [Online]. Available: ftp://ftp.usc.edu/pub/csinfo/tech-reports/papers/95-611.ps.Z.
[11] Che, H., Y. Tung, and Z. Wang, “Hierarchical Web caching systems: Modeling, design and experimental results,” IEEE J. Select. Areas Commun., Volume 20, Number 7, September 2002.
[12] Fan, L., P. Cao, and J. Almeida, “Summary Cache: A scalable wide-area Web cache sharing protocol,” in Proceedings of ACM SIGCOMM'98, Pages 254–265, August 31 – September 4, 1998.
[13] Feeley, M. J., W. E. Morgan, F. H. Pighin, A. R. Karlin, and H. M. Levy,“Implementing global memory management in a workstation cluster,” in Proceedings of 15th ACM Symp. Operating Systems Principles (SOSP), Pages 201–212, December 1995.
[14] Gadde, S., J. Chase, and M. Rabinovich, “Directory structures for scalable Internet caches,” Technical Report CS-1997-18, Dept. of Computer Science, Duke University, November 1997. [Online]. Available: ftp://ftp.cs.duke.edu/dist/techreport/1997/1997-18.ps.gz.
[15] Gadde, S., M. Rabinovich, and J. Chase, “Reduce, reuse, recycle: An approach to building large Internet caches,” in Proceedings of 6th Workshop on Hot Topics in Operating Systems (HotOS-VI), Pages 93–98, May 1997.
[16] Gadde, S., M. Rabinovich, and J. Chase, “A taste of Crispy Squid,” in Proceedings of 6th Workshop on Internet Server Performance (WISP), June 1998. [Online]. Available: http://www.cs.wisc.edu/~cao/WISP98/final-versions/chase.ps.
[17] Gettys, J., J. Mogul, H. Frystyk, L. Masinter, P. Leach, and T. Berners-Lee,“Hypertext transfer protocol – HTTP/1.1,” RFC 2616, June 1999.
[18] Gribble, S. and E. Brewer, “UC Berkeley home IP HTTP traces,” June 1997. [Online]. Available: http://ita.ee.lbl.gov/html/contrib/UCB.home-IP-HTTP.html.
[19] Karger, D., A. Sherman, A. Berkheimer, B. Bogstad, R. Dhanidina, K. Iwamoto, B. Kim, L. Matkins, and Y. Yerushalmi, “Web caching with consistent hashing,” in Proceedings of Eighth Int'l World Wide Web Conf. (WWW8), Pages 125–135, May 1999.
[20] Kelly, T. and J. C. Mogul, “Aliasing on the world wide web: Prevalence and performance implications,” in Proc. 11th World Wide Web Conf. (WWW11), Pages 281–292, May 2002.
[21] Kelly, T. and D. Reeves, “Optimal Web cache sizing: Scalable methods for exact solutions,” in Proceedings of 5th Int'l Web Caching and Content Delivery Workshop (WCW), Pages 163–173, May 2000.
[22] Lancellotti, R., F. Mazzoni, and M. Colajanni, “Hybrid cooperative schemes for scalable and stable performance ofWeb content delivery,” Computer Networks, Volume 49, Number 4, Pages 492–511, December 2005.
[23] Laoutaris, N., S. Syntila, and I. Stavrakakis, “Meta algorithms for hierarchical Web caches,” in Proceedings of 23rd IEEE Int'l Performance Computing and Communications Conf. (IPCCC), April 2004.
[24] Rabinovich,M., J. Chase, and S. Gadde, “Not all hits are created equal: Cooperative proxy caching over a wide-area network,” in Proceedings of 3rd Int'l WWW Caching Workshop (WCW), Pages 92–100, June 1998.
[25] Ross, K. W., “Hash routing for collections of shared Web caches,” IEEE Network, Volume 11, Number 6, Pages 37–44, November/December 1997.
[26] Rousskov, A., “On performance of caching proxies,” in Proceedings of 1998 ACM SIGMETRICS Int'l Conf. Measurement and Modeling of Computer Systems, Pages 272–273, June 1998.
[27] Rousskov, A. and D. Wessels, “Cache digests,” in Proceedings of 3rd Int'l WWW Caching Workshop (WCW), June 1998.
[28] Shedler, G. and C. Tung, “Locality in page reference strings,” SIAM Journal of Computing, Volume 1, Number 3, Pages 218–241, 1972.
[29] Sheu, T.-L. and C.-H. Yang, “A novel hierarchical cache architecture,” in Proceedings of 15th Int'l Conf. Information Networking (ICOIN), Pages 863–868, January 31 – February 2, 2001.
[30] Tewari, R., M. Dahlin, H. M. Vin, and J. S. Kay, “Design considerations for distributed caching on the Internet,” in Proceedings of 19th IEEE Int'l Conf. Distributed Computing Systems (ICDCS), Pages 273–284, May 31 – June 4, 1999.
[31] Valloppillil, V. and K. Ross, “Cache array routing procotol v1.0,” IETF Internet-Draft (work in progress), October 1997. [Online]. Available: http://www.ietf.org/internet-drafts/draft-vinod-carp-v1-02.txt.
[32] Wessels, D., “Squid Internet object cache,” [Online]. Available: http://www.squid-cache.org/.
[33] Wessels, D., “Sanitized access log,” July 1997. [Online]. Available: ftp://ircache.nlanr.net/Traces/, configuration files for the proxies are at http://ircache.nlanr.net/Cache/Configuration/.
[34] Wessels, D. and K. Claffy, “Internet cache protocol (ICP), version 2,” RFC 2186, September 1997.
[35] Wessels, D. and K. Claffy, “ICP and the Squid Web cache,” IEEE J. Select. Areas Commun., Volume 16, Number 3, Pages 345–357, April 1998.
[36] Wu, K. L. and P. S. Yu, “Latency-sensitive hashing for collaborative Web caching,” in Proceedings of Ninth Int'l World Wide Web Conf. (WWW9), Pages 633–644, May 2000.
[37] Zipf, G. K., Human Behavior and the Principle of Least Effort, Addison-Wesley Press, Cambridge, MA, 1949.