Subscribe
Issue No.10 - October (2010 vol.21)
pp: 1398-1411
Vinh Bui , University of New South Wales at Australian Defence Force Academy, Campbell
Weiping Zhu , University of New South Wales at Australian Defence Force Academy, Campbell
Alessio Botta , University of Napoli "Federico II", Napoli
Antonio Pescapé , University of Napoli "Federico II", Napoli
ABSTRACT
The use of multipath routing in overlay networks is a promising solution to improve performance and availability of Internet applications, without the replacement of the existing TCP/IP infrastructure. In this paper, we propose an approach to distribute data over multiple overlay paths that is able to improve Quality of Service (QoS) metrics, such as the data transfer time, loss, and throughput. By using the Imbedded Markov Chain technique, we demonstrate that the system under analysis, observed at specific instants, possesses the Markov property. We therefore cast the data distribution problem into the Markov Decision Process (MDP) framework, and design a computationally efficient algorithm named Online Policy Iteration (OPI), to solve the optimization problem on the fly. The proposed approach is applied to the problem of multipath data distribution in various wired/wireless network scenarios, with the objective of minimizing the data transfer time as well as the delay and losses. Through both intensive ns-2 simulations with data collected from real heterogeneous networks and experiments over real networks, we show the superior performance of the proposed traffic control mechanism in comparison with two classical schemes, that are Weighted Round Robin and Join the Shortest Queue.
INDEX TERMS
Multipath data transfer, Markov decision processes, overlay networks, heterogeneous networks.
CITATION
Vinh Bui, Weiping Zhu, Alessio Botta, Antonio Pescapé, "A Markovian Approach to Multipath Data Transfer in Overlay Networks", IEEE Transactions on Parallel & Distributed Systems, vol.21, no. 10, pp. 1398-1411, October 2010, doi:10.1109/TPDS.2010.13
REFERENCES
 [1] A. Akella et al., "A Measurement-Based Analysis of Multihoming," Proc. ACM SIGCOMM '03, pp. 353-364, Aug. 2003. [2] A. Akella et al., "A Comparison of Overlay Routing and Multihoming Route Control," Proc. ACM SIGCOMM '04, pp. 93-106, Aug. 2004. [3] E. Altman, "Applications of Markov Decision Processes in Communication Networks," Handbook of Markov Decision Processes: Methods and Applications, Kluwer, 2002. [4] D. Andersen et al., "Resilient Overlay Networks," ACM SIGOPS Operating Systems Rev., vol. 35, no. 5, pp. 131-145, Dec. 2001. [5] D.G. Andersen, "Improving End-to-End Availability Using Overlay Networks," PhD, Massachusetts Inst. of Tech nology, Feb. 2005. [6] C. Barakat et al., "Modeling Internet Backbone Traffic at the Flow Level," IEEE Trans. Signal Process., vol. 51, no. 8, pp. 2111-2124, Aug. 2003. [7] A. Botta et al., "Identification of Network Bricks in Heterogeneous Scenarios," Computer Networks, vol. 52, no. 15, pp. 2975-2987, 2008. [8] V. Bui et al., "Improving Multipath Live Streaming Performance with Markov Decision Processes," Proc. IEEE Int'l Symp. Comm. and Information Technologies (ISCIT '07), Oct. 2007. [9] V. Bui et al., "Modelling Internet End-to-End Loss Behaviours: A New Approach," Proc. Asia Modelling Symp. '07, Mar. 2007. [10] V. Bui et al., "A Game Theoretic Framework for Multipath Optimal Data Transfer in Multiuser Overlay Networks," Proc. IEEE Int'l Conf. Comm. (ICC '08), May 2008. [11] V. Bui et al., "An Mdp-Based Approach for Multipath Data Transmission over Wireless Networks," Proc. IEEE Int'l Conf. Comm. (ICC '08), May 2008. [12] C. Cetinkaya et al., "Opportunistic Traffic Scheduling Over Multiple Network Paths," Proc. IEEE INFOCOM, pp. 1928-1937, Mar. 2004. [13] J. Chen et al., "An Efficient Multipath Forwarding Method," Proc. IEEE INFOCOM, pp. 1418-1425, Mar. 1998. [14] W.C. Cheng et al., "Large-scale Data Collection: A Coordinated Approach," Proc. IEEE INFOCOM '03, pp. 218-228, Mar. 2003. [15] D.R. Cox, Renewal Theory. Methuen & Co., 1962. [16] S. Ganguly et al., "Fast Replication in Content Distribution Overlays," Proc. IEEE INFOCOM, pp. 2246-2256, Mar. 2005. [17] Y. Hasegawa et al., "Improved Data Distribution for Multipath Tcp Communication," Proc. IEEE Global Telecomm. Conf. (GLOBECOM), 2005. [18] J. He and J. Rexford, "Toward Internet-Wide Multipath Routing," IEEE Network Magazine, vol. 22, no. 2, pp. 16-21, Mar./Apr. 2008. [19] J.R. Iyengar, P.D. Amer, and R. Stewart, "Concurrent Multipath Transfer Using SCTP Multihoming over Independent End-to-End Paths," IEEE/ACM Trans. Networking, vol. 14, no. 5, pp. 951-964, OCt. 2006. [20] J.R. Iyengar et al., "Concurrent Multipath Transfer Using Transport Layer Multihoming: Performance Under Varying Bandwidth Proportions," Proc. IEEE Military Comm. Conf. (MILCOM), pp. 238-244, Oct. 2004. [21] D. Johnson et al., "DSR: The Dynamic Source Routing Protocol for Multihop Wireless Ad Hoc Networks," Ad Hoc Networking, ch. 5, pp. 139-172, Addison-Wesley Longman Publishing Co., Inc., 2001. [22] D. Jurca et al., "Video Packet Selection and Scheduling for Multipath Streaming," IEEE Trans. Multimedia, vol. 9, no. 3, pp. 629-641. Apr. 2007. [23] T. Karagiannis et al., "A Nonstationary Poisson View of Internet Traffic," Proc. IEEE INFOCOM, pp. 177-187, Jan. 2004. [24] L. Kleinrock, Queueing Systems Volume 1: Theory. John Wiley & Son, 1975. [25] A. Konrad et al., "Choosing an Accurate Network Path Model," Proc. ACM SIGMETRICS '03, pp. 314-315, Jun. 2003. [26] S.-J. Lee and M. Gerla, "Split Multipath Routing with Maximally Disjoint Paths in Ad Hoc Networks," Proc. IEEE Int'l Conf. Comm. (ICC '01), 2001. [27] J.Y.-T. Leung, Handbook of Scheduling: Algorithms, Models, and Performance Analysis. Chapman and Hall/CRC, 2004. [28] Y. Liang, J. Apostolopoulos, and B. Girod, "Analysis of Packet Loss for Compressed Video: Does Burst-Length Matter?" Proc. IEEE Int'l Conf. Acoustics, Speech, and Signal Processing, 2003. [29] J. Little, "A Proof of the Queueing Formula $l=\lambda w$ ," Operations Research, vol. 9, pp. 383-387, 1961. [30] X. Liu et al., "Opportunistic Transmission Scheduling with Resource-Sharing Constraints in Wireless Networks," IEEE J. Selected Areas Comm., vol. 19, no. 10, pp. 2053-2064, Oct. 2001. [31] Z. Ma et al., "A New Multi-Path Selection Scheme for Video Streaming on Overlay Networks," Proc. IEEE Int'l Conf. Comm. (ICC), 2004. [32] S. Pederson and M. Johnson, "Estimating Model Discrepancy," Technometrics, vol. 32, no. 3, pp. 305-314, Aug. 1990. [33] A. Pescapé et al., "End-to-End Packet-Channel Bayesian Model Applied to Heterogeneous Wireless Networks," Proc. IEEE Global Telecomm. Conf. (GLOBECOM '05), vol. 1, Dec. 2005. [34] M.L. Puterman, Markov Decision Processes: Discrete Stochastic Dynamic Programming. John Wiley and Sons, 1994. [35] R.S. Sutto et al., Reinforcement Learning: An Introduction. The MIT Press, 1999. [36] S. Tao et al., "Exploring the Performance Benefits of End-to-End Path Switching," Proc. IEEE Int'l Conf. Network Protocols (ICNP '04), pp. 304-315, Oct. 2004. [37] R. Teixeira, K. Marzullo, S. Savage, and G.M. Voelker, "In Search of Path Diversity in ISP Networks," Proc. ACM Internet Measurement Conf. (IMC '03), 2003. [38] E. Vergetis et al., "Packet-Level Diversity—From Theory to Practice: An 802.11-Based Experimental Investigation," Proc. ACM MobiCom '06, 2006. [39] L. Vu et al., "Measurement and Modeling a Large-Scale Overlay for Multimedia Streaming," Proc. ICST Int'l Conf. Heterogeneous Networking for Quality, Reliability, Security and Robustness (QShine '07), 2007. [40] S. Vutukury et al., "MPATH: A Loop-Free Multipath Routing Algorithm," J. Microprocessors and Microsystems, vol. 24, pp. 319-327, 2000. [41] B. Wang et al., "Multipath Overlay Data Transfer," technical report, Univ. of Massachusetts Amherst, 2005. [42] B. Wang et al., "Application-Layer Multipath Data Transfer via TCP: Schemes and Performance Tradeoffs," J. Performance Evaluation, vol. 64, pp. 965-977, 2007. [43] Y. Zhang et al., "On the Constancy of Internet Path Properties," Proc. ACM SIGCOMM Workshop Internet Measurement, 2001.