Providing proportional TCP performance by fixed-point approximations over bandwidth on demand satellite networks

Article English OPEN
Chai, Wei Koong ; Karaliopoulos, M. ; Pavlou, G. (2009)
  • Related identifiers: doi: 10.1109/TWC.2009.071395
  • Subject:
    acm: ComputerSystemsOrganization_COMPUTER-COMMUNICATIONNETWORKS
    arxiv: Computer Science::Performance | Computer Science::Networking and Internet Architecture | Computer Science::Operating Systems

In this paper we focus on the provision of propor-\ud tional class-based service differentiation to transmission control protocol (TCP) flows in the context of bandwidth on demand(BoD) split-TCP geostationary (GEO) satellite networks. Our approach involves the joint configuration of TCP-Performance Enhancing Proxy (TCP-PEP) agents at the transport layer and the scheduling algorithm controlling the resource allocation at the Medium Access Control (MAC) layer. We show that the two differentiation mechanisms exhibit complementary behavior in achieving the desired differentiation throughout the traffic load space: the TCP-PEPs control differentiation at low and medium system utilization, whereas the MAC scheduler becomes the dominant differentiation factor under high traffic load. The main challenge for the satellite operator is to appropriately configure those two mechanisms to achieve a specific differentiation target for the different classes of TCP flows. To this end, we propose a fixed-point framework to analytically approximate the achieved differentiated TCP performance. We validate the predictive capacity of our analytical method via simulations and show that our approximations closely match the performance of different classes of TCP flows under various scenarios for the\ud network traffic load and configuration of the MAC scheduler\ud and TCP-PEP agent. Satellite network operators could use our\ud approximations as an analytical tool to tune their networks
  • References (30)
    30 references, page 1 of 3

    [1] H. Skinnemoen and H. Tork, “Standardization activities within broadband satellite multimedia," in Proc. IEEE ICC 2002, vol. 5, New York, NY, 2002, pp. 3010-3014.

    [2] R. Braden, D. Clark and S. Shenker, “Integrated services in the internet architecture: an overview," RFC 1633, June 1994.

    [3] S. Blake, D. Black, M. Carlson, E. Davies, Z. Wang and W. Weiss, “An architecture for differentiated service," RFC 2475, Dec. 1998.

    [4] A. Iera, A. Molinaro and S. Marano, “IP with QoS guarantees via Geo satellite channels: performance issues," IEEE Personal Commun., vol. 8, no. 3, pp. 14-19, June 2001.

    [5] L. S. Ronga, T. Pecorella, E. Del Re and R. Fantacci, “A gateway architecture for IP satellite networks with dynamic resource management and DiffServ QoS provision," Int'l J. of Satell. Commun. Networking, no. 21, pp. 351-366, 2003.

    [6] N. Iuoras and T. Le-Ngoc, “Dynamic capacity allocation for quality-ofservice support in IP-based satellite networks," IEEE Wireless Commun. Mag., vol. 12, issue 5, pp. 14-20, Oct. 2005.

    [7] C. Dovrolis, D. Stiliadis and P. Ramanathan, “Proportional differentiated services: delay differentiation and packet scheduling," IEEE/ACM Trans. Networking, vol. 10, no. 1, pp. 12-26, Feb. 2002.

    [8] M. Welzl and M. Muhlhauser, “Scalability and quality of service: a trade-off?", IEEE Commun. Mag., pp. 32-36, June 2003.

    [9] K. Wang, “Quality of service assurances in multihop wireless networks," PhD. Dissertation, University of Wisconsin-Madison, 2003.

    [10] C-W. Tan, M. Gurusamy and J. C-S. Lui, “Achieving multi-class service differentiation in WDM optical burst switching networks: A probabilistic preemptive burst segmentation scheme," IEEE J. Select. Areas Commun., vol. 24, no. 12, Dec. 2006.

  • Metrics
    No metrics available
Share - Bookmark