臺灣博碩士論文加值系統

　　TCP splicing是一種不了解使用者請求內容的第四層交換器技術，也就是說，第四層交換器無法使用應用層的資訊做封包轉送。在本文中，我們提出一個TCP splicing的延伸技術，稱為FPS，FPS使得TCP splicing得以使用應用層的資訊做負載平衡。在FPS中，一個使用者的連線將在FPS交換器與所有的後端伺服器之間預先建立連線；除此之外，FPS利用在核心層中擷取應用層的資訊以減少資料在使用者層與核心層之間的複製時間。我們從LVS中取得TCPSP並且修改TCPSP以實作FPS，而LVS則是利用Linux Netfilter實作。在叢集式的環境中，實驗存取動態網頁的結果顯示：FPS在產能與延遲時間方面分別比TCPSP優良四倍與五倍。

　　TCP splicing is a content-blind layer-4 switch technique that prevents the switches from using the application layer information for forwarding decisions. In this paper, we proposed a new technique called fully pre-splicing (FPS) which is an extension of TCP splicing. FPS extends TCP splicing to support content-aware load balancing algorithms. In FPS, one connection is pre-established between the FPS switch and each back-end server. In addition, FPS extracts the application information in kernel space for eliminating the cost of moving data between user and kernel spaces. We implemented the proposed FPS by modifying the TCPSP project of LVS using Linux Netfilter. The performance results show that the throughput and latency of accessing dynamic web pages in clustered web servers using FPS switch is four times and five times better than that using TCPSP, respectively.

Chapter 1 Introduction 1
Chapter 2 Related Work 4
2.1 Two-way architecture 5
2.1.1 TCP gateway 6
2.1.2 TCP Splicing [5] 7
2.1.3 TCP Tap [7] 8
2.1.4 Asymmetric TCP Splicing [9] 10
2.1.5 KNITS [10] 11
2.1.6 DHTTP [11] 11
2.2 One-way architecture 13
2.2.1 TCP connection hop [12] 14
2.2.2 TCP handoff [6] 15
2.2.3 DRWS [13] 16
2.2.4 Socket Cloning [14] 17
Chapter 3 Fully Pre-Splicing (FPS) 21
3.1 Architecture 21
3.2 The Mechanism of FPS 30
3.3 A complete example of the FPS Mechanism 33
3.4 Implementation 35
Chapter 4 Performance Evaluation 39
4.1 Benchmark 39
4.2 Overhead Results 40
4.3 TCPSP versus FPS 43
4.4 SSA versus FPS 49
Chapter 5 Conclusion and Future work 51
5.1 Conclusion 51
5.2 Future work 51
References 52

[1]E. Casalicchio and M. Colajanni. A client-aware dispatching algorithm for web clusters providing multiple services. In Proceedings of the 10th International World Wide web Conference, pages 535-544, Hong Kong, May 2001.
[2]V. S. Pai, M. Aron, G. Banga, M. Svendsen, P. Druschel, W. Zwaenepoel, and E. M. Nahum. “Locality-Aware Request Distribution in Cluster-based Network Servers”, In Proceedings of the 8th ACM Conference on Architectural Support for Programming Languages and Operating Systems, pages 205-216, San Jose, CA, Oct. 1998.
[3]L. Cherkasova and M. Karlsson, “Scalable web server cluster design with workload-aware request distribution strategy WARD”, 3rd International Workshop on Advanced Issues of E-Commerce and web-Based Information Systems, WECWIS 2001, pp. 212-221.
[4]L. Cherkasova. “FLEX: Load balancing and management strategy for scalable web hosting service”, In Proceedings of the Fifth International Symposium on Computers and Communications (ISCC’00), pages 8-13, July 2000.
[5]Maltz, D., and Bhagwat, P. “TCP Splicing for Application Layer Proxy Performance”, IBM Research Report RC 21139(03/17/98) (March 1998).
[6]M. Aron, D. Sanders, P. Druschel, and W. Zwaenepoel. “Scalable Content-aware Request Distribution in Cluster-based Network Servers”, In Proceedings of the USENIX 2000 Annual Technical Conference, San Diego, CA, June 2000. http://citeseer.nj.nec.com/aron00scalable.html
[7]Maltz, D. A., and Bhagwat, P. “Improving HTTP Caching Proxy Performance with TCP Tap”, IBM Research Report RC 21147 (1998).
[8]Ariel Cohen, S. R., and Slye, H. “On the Performance of TCP Splicing for URL-aware Redirection”, Second USENIX Symposium on Internet Technologies and Systems, Boulder, CO, Oct. 1999.
[9]Masayoshi Kobayashi and Tutomu Murase. “Asymmetric TCP Splicing for Content-Based Switches”, Networking Research Laboratories, NEC Corp.(2002)
[10]A. E. Papathanasiou and E. V. Hensbergen, “KNITS: Switch-based connection handoff”, Proc. of IEEE Infocom, New York, USA, June 23-27, 2002.
[11]Michael Rabinovich and Hua Wang, “DHTTP: An Efficient and Cache-Friendly Transfer Protocol for web Traffic”, INFOCOM, 1597-1606, 2001.
[12]RESONATE. 2002. http://www.resonate.com/.
[13]Ying-Dar Lin, Ping-Tsai Tsai, Po-Ching Lin, Ching-Ming Tien, “Direct web Switch Routing with State Migration, TCP Masquerade, and Cookie name Rewriting”, Globecom, San Francisco, December 2003.
[14]Yiu-Fai Sit, Cho-Li Wang, Francis Lau. “Socket Cloning for Cluster-Based web Servers”, IEEE Fourth International Conference on Cluster Computing (CLUSTER 2002) chicago, USA
[15]C. Yang, M. Luo, “Efficient Support for Content-Based Routing in web Server Clusters”, Proceedings of USITS’99-2nd USENIX Symposium on Internet Technologies & Systems, Boulder, Colorado, USA, October 11-14, 1999.
[16]TCP Splicing project (TCPSP), http://www.linuxvirtualserver.org/software/tcpsp/index.html.
[17]Linux Virtual Server Project, 2002, http://www.linuxvirtualserver.org/.
[18]Netfilter, http://www.netfilter.org/.
[19]E. Casalicchio and M. Colajanni, “Scalable web cluster with static and dynamic contents”, Proceedings of IEEE Int’l Conf. on Cluster Computing, pp. 170-177, Chemnitz, Germany, Dec. 2000
[20]Du Zeng-Kai and Ju Jiu-Bin, “A completely distributed architecture for cluster-based web servers”, Proceedings of the Fourth International Conference on PDCAT’2003 , page 483-486, Aug. 2003.
[21]M. Andreolini, M. Colajanni, M. Nuccio, “Kernel-based web switches providing content-aware routing”, Proc. of 2nd IEEE Int’l Symposium on Network Computing and Application (NCA’03), Cambridge, MA, Apr. 2003.
[22]M. Di Santo, N. Ranaldo, E.Zimeo, “Kernel Implementations of Locality-Aware Dispatching Techniques for web Server Clusters”, IEEE International Conference on Cluster Computing (CLUSTER’03), 2003.
[23]Thiemo Voigt and Per Gunningberg, “Kernel-based Control of Persistent web Server Connections”, ACM Performance Evaluation Review, pp. 20-25, Sep 2001.
[24]L. Aversa and A. Bestavro, “Load balancing a cluster of web servers using Distributed Packet Rewriting”, In Proceedings of the 19th IEEE International Performance, Computing, and Communication Conference, pages 24-29, Phoenix, AZ, Feb. 2000.
[25]G. Apostolopoulos, D. Aubespin, V. Peris, P. Pradhan, and D. Saha. “Design, implementation and Performance of a Content-based switch”, In Proceedings of the 19th IEEE International Conference on Computer Communications (INFOCOM 2000), pages 1117-1126, Tel-Aviv, Israel, Mar. 2000.
[26]V. Cardellini, E. Casalicchio, M. Colajanni, and P. Yu. “The state of the art in locally distributed web-server system”, ACM Computing Surveys, 34(2), June 2002.
[27]ArrowPoint Communications Inc. (aquired by Cisco), “A Comparative Analysis of Web Switching Architectures”, http://www.arrowpoint.com/solutions/whitepapers/wsarchv6.html, 1998.
[28]G. Apostolopoulos and D. Aubespin and V. Peris and P. Pradhan and D. Saha, "Design, Implementation and Performance of a Content-Based Switch", http://citeseer.nj.nec.com/206701.html.
[29]Netscape Communications Corporation, “PERSISTENT CLIENT STATE HTTP COOKIES - Preliminary Specification”, http://home.netscape.com/newsref/std/cookie_spec.html.
[30]Nortel Networks, Alteon 180 series, http://www.nortelnetworks.com/products/01/alt180/.
[31]Cisco, isco CSS 11000 series content services switches, http://www.cisco.com/warp/public/cc/pd/si/11000/.
[32]Foundry Networks, BigIron, http://www.foundrynet.com/products/l3backbone/ bigiron/index.html