A Greedy Algorithm for Capacity-Constrained Surrogate Placement in CDNs

One major factor that heavily affects the performance of a content distribution network (CDN) is placement of the surrogates. Previous works take a network-centric approach and consider only the network traffic. In this paper, we propose solutions to optimal surrogate placement, taking into consideration both network latency and capacity constraints on the surrogates. For CDNs with a tree topology, an efficient and effective greedy algorithm is proposed which minimizes network traffic while at the same time maximizing system throughput. Simulation results show that the greedy algorithm is far better than the existing optimal placement scheme that makes decisions based solely on network traffic. This suggests that capacity constraints on surrogates or server bottlenecks should be considered when determining surrogate placement, especially when the capacities of CDN servers are limited.

[1]  I. Lazar,et al.  Exploring content delivery networking , 2001 .

[2]  Rahul Simha,et al.  A Microeconomic Approach to Optimal Resource Allocation in Distributed Computer Systems , 1989, IEEE Trans. Computers.

[3]  Maria Kihl,et al.  Web server performance modeling using an M/G/1/K*PS queue , 2003, 10th International Conference on Telecommunications, 2003. ICT 2003..

[4]  Xueyan Tang,et al.  Coordinated En-Route Web Caching , 2002, IEEE Trans. Computers.

[5]  Lili Qiu,et al.  On the placement of Web server replicas , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[6]  Fabián A. Chudak,et al.  Improved approximation algorithms for a capacitated facility location problem , 1999, SODA '99.

[7]  Anirban Mahanti,et al.  Traffic analysis of a Web proxy caching hierarchy , 2000 .

[8]  Lawrence W. Dowdy,et al.  Comparative Models of the File Assignment Problem , 1982, CSUR.

[9]  Li Fan,et al.  Web caching and Zipf-like distributions: evidence and implications , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).

[10]  P. Krishnan,et al.  The cache location problem , 2000, TNET.

[11]  Pascal Felber,et al.  Self-organization in Cooperative Content Distribution Networks , 2005, Fourth IEEE International Symposium on Network Computing and Applications.

[12]  Brad Cain,et al.  A Model for Content Internetworking (CDI) , 2003, RFC.

[13]  Ming T. Liu,et al.  Optimization of performance gain in content distribution networks with server replicas , 2003, 2003 Symposium on Applications and the Internet, 2003. Proceedings..

[14]  Xiaohua Jia,et al.  Placement of read-write Web proxies on the Internet , 2001, Proceedings 21st International Conference on Distributed Computing Systems.

[15]  Bo Li,et al.  On the optimal placement of web proxies in the Internet , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).

[16]  Ouri Wolfson,et al.  The multicast policy and its relationship to replicated data placement , 1991, TODS.

[17]  Yuval Shavitt,et al.  Constrained mirror placement on the Internet , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[18]  Arun Venkataramani,et al.  Bandwidth constrained placement in a WAN , 2001, PODC '01.

[19]  Jianliang Xu,et al.  Placement problems for transparent data replication proxy services , 2002, IEEE J. Sel. Areas Commun..

[20]  Guerney D. H. Hunt,et al.  Network Dispatcher: A Connection Router for Scalable Internet Services , 1998, Comput. Networks.

[21]  Abdelsalam Heddaya,et al.  WebWave: globally load balanced fully distributed caching of hot published documents , 1997, Proceedings of 17th International Conference on Distributed Computing Systems.