A novel hybrid probing technique for end-to-end available bandwidth estimation

The information of available bandwidth on an end-to-end path is important for various network applications, and several probing methods have been proposed to estimate it in recent years. However, previous methods are either based on fluid model or are only partially suitable for bursty real internet cross traffic; and the accuracy of their estimation degrades at different extents in multi-hop situations. Moreover, all previous PGM (Probing Gap Model) based methods require the knowledge of bottleneck link capacity, which may not be available in practice. In this paper, we extend the analysis of queuing behavior of probing packets from single-hop scenarios to multi-hop scenarios and propose a novel hybrid probing technique, called PATHCOS++, which integrates the advantages of both PRM (Probing Rate Model) and PGM based methods, to estimate the end-to-end available bandwidth. Unlike previous works, PATHCOS++ does not make fluid cross traffic assumption and does not require the information about bottleneck link capacity. Simulation results show that PATHCOS++ is quite efficient and provides end-to-end available bandwidth estimation that is significantly more accurate than current state-of-the-art techniques do. The accuracy of PATHCOS++ is nearly unaffected when there are multiple congestible links.

[1]  Manish Jain,et al.  End-to-end available bandwidth: measurement methodology, dynamics, and relation with TCP throughput , 2003, IEEE/ACM Trans. Netw..

[2]  D. Loguinov,et al.  A Stochastic Foundation of Available Bandwidth Estimation: Multi-Hop Analysis , 2008, IEEE/ACM Transactions on Networking.

[3]  M. Frans Kaashoek,et al.  A measurement study of available bandwidth estimation tools , 2003, IMC '03.

[4]  Shivendra S. Panwar,et al.  On optimal partitioning of realtime traffic over multiple paths , 2005, Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies..

[5]  George Yang,et al.  Network Characterization Service (NCS) , 2001, Proceedings 10th IEEE International Symposium on High Performance Distributed Computing.

[6]  Robert Tappan Morris,et al.  Resilient overlay networks , 2001, SOSP.

[7]  Yuguang Fang,et al.  Admission Control Based on Available Bandwidth Estimation for Wireless Mesh Networks , 2009, IEEE Transactions on Vehicular Technology.

[8]  Dmitri Loguinov,et al.  What signals do packet-pair dispersions carry? , 2005, Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies..

[9]  Johan M. Karlsson,et al.  Real-time available-bandwidth estimation using filtering and change detection , 2009, Comput. Networks.

[10]  Zhao Wen-tao,et al.  Efficient available bandwidth estimation for network paths , 2008 .

[11]  Jia Wang,et al.  A measurement study of Internet bottlenecks , 2005, Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies..

[12]  Richard G. Baraniuk,et al.  pathChirp: Efficient available bandwidth estimation for network paths , 2003 .

[13]  Shahrokh Valaee,et al.  A System-Theoretic Approach to Bandwidth Estimation , 2008, IEEE/ACM Transactions on Networking.

[14]  Mats Björkman,et al.  A new end-to-end probing and analysis method for estimating bandwidth bottlenecks , 2000, Globecom '00 - IEEE. Global Telecommunications Conference. Conference Record (Cat. No.00CH37137).

[15]  Alok Shriram,et al.  Empirical Evaluation of Techniques for Measuring Available Bandwidth , 2007, IEEE INFOCOM 2007 - 26th IEEE International Conference on Computer Communications.

[16]  Van Jacobson,et al.  Congestion avoidance and control , 1988, SIGCOMM '88.

[17]  Srinivasan Keshav,et al.  A control-theoretic approach to flow control , 1991, SIGCOMM '91.

[18]  Josep Mangues-Bafalluy,et al.  Impact of transient CSMA/CA access delays on active bandwidth measurements , 2009, IMC '09.

[19]  Mostafa H. Ammar,et al.  Poisson versus periodic path probing (or, does PASTA matter?) , 2005, IMC '05.

[20]  Jennifer C. Hou,et al.  On exploiting long range dependence of network traffic in measuring cross traffic on an end-to-end basis , 2003, IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37428).

[21]  Sridhar Machiraju,et al.  The role of PASTA in network measurement , 2006, SIGCOMM 2006.

[22]  Hari Balakrishnan,et al.  Resilient overlay networks , 2001, SOSP.

[23]  Mark Crovella,et al.  Measuring Bottleneck Link Speed in Packet-Switched Networks , 1996, Perform. Evaluation.

[24]  Anees Shaikh,et al.  An empirical evaluation of wide-area internet bottlenecks , 2003, SIGMETRICS '03.

[25]  Anees Shaikh,et al.  An empirical evaluation of wide-area internet bottlenecks , 2003 .

[26]  Miguel A. Labrador,et al.  Traceband: A fast, low overhead and accurate tool for available bandwidth estimation and monitoring , 2010, Comput. Networks.

[27]  Peter Steenkiste,et al.  Improving TCP startup performance using active measurements: algorithm and evaluation , 2003, 11th IEEE International Conference on Network Protocols, 2003. Proceedings..

[28]  Richard G. Baraniuk,et al.  Multifractal Cross-Traffic Estimation , 2000 .

[29]  M. Y. Sanadidi,et al.  The probe gap model can underestimate the available bandwidth of multihop paths , 2006, CCRV.

[30]  Parameswaran Ramanathan,et al.  What do packet dispersion techniques measure? , 2001, Proceedings IEEE INFOCOM 2001. Conference on Computer Communications. Twentieth Annual Joint Conference of the IEEE Computer and Communications Society (Cat. No.01CH37213).

[31]  D. Loguinov,et al.  A Queueing-Theoretic Foundation of Available Bandwidth Estimation: Single-Hop Analysis , 2007, IEEE/ACM Transactions on Networking.

[32]  Parameswaran Ramanathan,et al.  Packet-dispersion techniques and a capacity-estimation methodology , 2004, IEEE/ACM Transactions on Networking.

[33]  Manish Jain,et al.  End-to-end available bandwidth: measurement methodology, dynamics, and relation with TCP throughput , 2003, TNET.