On Using Circuit-Switched Networks for File Transfers

High-speed optical circuit-switched networks are being deployed to support highly demanding applications in the eScience research community. File transfers were identified as an ideal application for high-speed circuits because such transfers can utilize any rate, limited only by end-host capabilities. This paper addresses the question of how a circuit-switched network can be shared efficiently amongst many users for file transfers. Specifically, we study the question of how much bandwidth should be allocated per file transfer so that the system performance averaged across all flows is optimal in terms of mean response time. We explore a heterogeneous bandwidth-allocation scheme, in which bandwidth is allocated based on file sizes, as well as a homogeneous scheme, in which all transfers are allocated the same amount of bandwidth. Our analytical results show that bandwidth allocation should be based on loading conditions. Under low loads, we recommend a homogeneous scheme and each call should be allocated a high-bandwidth circuit. However, under high loads, small files should be allocated low-bandwidth circuits and large files high-bandwidth circuits; this heterogeneous bandwidth-allocation scheme results in a one-to-three orders of magnitude reduction in mean response time.

[1]  C. Laws,et al.  Optimization via trunk reservation in single resource loss systems under heavy traffic , 1997 .

[2]  Pablo Molinero-Fernández,et al.  Systems with multiple servers under heavy-tailed workloads , 2005, Perform. Evaluation.

[3]  Terry Moore,et al.  An end-to-end approach to globally scalable network storage , 2002, SIGCOMM 2002.

[4]  Charles R. Kalmanek,et al.  A retrospective view of ATM , 2002, CCRV.

[5]  Scott Shenker,et al.  Overcoming the Internet impasse through virtualization , 2005, Computer.

[6]  R. Callon.  Internetwork protocol , 1983, Proceedings of the IEEE.

[7]  Gigi Karmous-Edwards,et al.  An optical control plane for the grid community: opportunities, challenges, and vision , 2006, IEEE Communications Magazine.

[8]  Robert E. Kahn,et al.  A Protocol for Packet Network Intercommunication , 1974 .

[9]  Sally Floyd,et al.  Wide area traffic: the failure of Poisson modeling , 1995, TNET.

[10]  Hiroshi Kitamura,et al.  A SOCKS-based IPv6/IPv4 Gateway Mechanism , 2001, RFC.

[11]  Raj Jain,et al.  The art of computer systems performance analysis - techniques for experimental design, measurement, simulation, and modeling , 1991, Wiley professional computing.

[12]  Malathi Veeraraghavan,et al.  A study of bandwidth-sharing mechanisms in connection-oriented networks , 2008 .

[13]  Kin K. Leung,et al.  An inversion algorithm for loss networks with state-dependent rates , 1995, Proceedings of INFOCOM'95.

[14]  Chase Qishi Wu,et al.  CHEETAH: circuit-switched high-speed end-to-end transport architecture testbed , 2005, IEEE Communications Magazine.

[15]  Cheng-Zhong Xu,et al.  Robust processing rate allocation for proportional slowdown differentiation on Internet servers , 2004, IEEE Transactions on Computers.

[16]  Myron Hlynka,et al.  Queueing Networks and Markov Chains (Modeling and Performance Evaluation With Computer Science Applications) , 2007, Technometrics.

[17]  Pablo Molinero-Fernández,et al.  TCP Switching: Exposing Circuits to IP , 2002, IEEE Micro.

[18]  Azer Bestavros,et al.  Self-similarity in World Wide Web traffic: evidence and possible causes , 1996, SIGMETRICS '96.

[19]  Benjamin Avi-Itzhak,et al.  SQF: A slowdown queueing fairness measure , 2007, Perform. Evaluation.

[20]  Allen B. Downey,et al.  A parallel workload model and its implications for processor allocation , 1996, Proceedings. The Sixth IEEE International Symposium on High Performance Distributed Computing (Cat. No.97TB100183).

[21]  T. J. Shepard,et al.  TCP/IP performance over satellite links , 1997, IEEE Netw..

[22]  Himabindu Pucha,et al.  Overlay TCP: Ending End-to-End Transport for Higher Throughput , 2005 .

[23]  Donald F. Towsley,et al.  Modeling TCP throughput: a simple model and its empirical validation , 1998, SIGCOMM '98.

[24]  Matthew Mathis,et al.  The macroscopic behavior of the TCP congestion avoidance algorithm , 1997, CCRV.

[25]  Mor Harchol-Balter.  Task assignment with unknown duration , 2002, JACM.

[26]  G. J. A. Stern,et al.  Queueing Systems, Volume 2: Computer Applications , 1976 .

[27]  V. Jacobson,et al.  Congestion avoidance and control , 1988, CCRV.

[28]  Malathi Veeraraghavan,et al.  CAM03-1: On the Suitability of Applications for GMPLS Networks , 2006, IEEE Globecom 2006.

[29]  Eric Hoffman,et al.  Flow labelled IP over ATM: design and rationale , 2006, CCRV.

[30]  Malathi Veeraraghavan,et al.  An overlay approach for enabling access to dynamically shared backbone GMPLS networks , 2007, 2007 16th International Conference on Computer Communications and Networks.

[31]  Jon Postel,et al.  Internet Protocol , 1981, RFC.

[32]  Qiang Song,et al.  Implementation of a GMPLS-based Network with End Host Initiated Signaling , 2006, 2006 IEEE International Conference on Communications.

[33]  Duane Wessels,et al.  High‐performance benchmarking with Web Polygraph , 2004, Softw. Pract. Exp..

[34]  Greg Minshall,et al.  Flow labelled IP: a connectionless approach to ATM , 1996, Proceedings of IEEE INFOCOM '96. Conference on Computer Communications.

[35]  Malathi Veeraraghavan,et al.  Analysis and design of book-ahead bandwidth-sharing mechanisms , 2008, IEEE Transactions on Communications.

[36]  Malathi Veeraraghavan,et al.  A reconfigurable ethernet/SONET circuit-based metro network architecture , 2004, IEEE Journal on Selected Areas in Communications.

[37]  Peter B. Danzig,et al.  Comparison of measurement-based admission control algorithms for controlled-load service , 1997, Proceedings of INFOCOM '97.

[38]  Wu-chun Feng,et al.  CHEETAH: circuit-switched high-speed end-to-end transport architecture , 2003, OptiComm: Optical Networking and Communications Conference.

[39]  Eric Mannie,et al.  Generalized Multi-Protocol Label Switching (GMPLS) Architecture , 2004, RFC.

[40]  Jerome H. Saltzer,et al.  End-to-end arguments in system design , 1984, TOCS.

[41]  Mischa Schwartz,et al.  Telecommunication networks: protocols, modeling and analysis , 1986 .

[42]  Injong Rhee,et al.  Binary increase congestion control (BIC) for fast long-distance networks , 2004, IEEE INFOCOM 2004.

[43]  Nick McKeown,et al.  A simulation study of IP switching , 1997, SIGCOMM '97.

[44]  David D. Clark,et al.  The design philosophy of the DARPA internet protocols , 1988, SIGCOMM '88.

[45]  Malathi Veeraraghavan,et al.  Experiences in implementing an experimental wide-area GMPLS network , 2007, IEEE Journal on Selected Areas in Communications.

[46]  Dimitri P. Bertsekas,et al.  Data Networks , 1986 .

[47]  A. Alles,et al.  ATM Internetworking , 1995 .

[48]  Greg Minshall,et al.  Ip Switching: Atm under Ip * , 1998 .

[49]  Daniel O. Awduche,et al.  IP over Optical Networks: A Framework , 2004, RFC.

[50]  Chitra Dorai,et al.  Extracting motion annotations from MPEG-2 compressed video for HDTV content management applications , 1999, Proceedings IEEE International Conference on Multimedia Computing and Systems.

[51]  Malathi Veeraraghavan,et al.  A Transport Protocol for Dedicated End-to-End Circuits , 2006, 2006 IEEE International Conference on Communications.

[52]  Stefan Savage,et al.  Modeling TCP latency , 2000, Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064).