On the Potential of Flow-Based Routing in Multihomed Environments

The data rates provisioned by broadband Internet access con ne tions continue to fall short of the requirements posed by emerging applications. Y et the potential of statistical multiplexing of the last mile broadband connections r emains unexploited even as the average utilization of these connections remains low. D espite recent work in this area [15, 20], two key questions remain unanswered: a) What i s the attainable benefit of broadband access sharing? and b) How much of this benefit is realizable given real-world constraints? In this work we quantify the attain able benefit of a multihomed broadband access environment by proposing and evaluating s everal flow-based access sharing policies using a custom flow-based simulator. We the n analyze how much of the performance benefit is lost due to real-world constraint s by migrating from simulations to a test-lab environment employing a wireless networ k. Our results show that in today’s broadband Internet access scenarios, a significant reduction in download times by up to a factor of 3 is achievable.

[1]  J.P. Singh,et al.  An Experimental Evaluation of Urban Networking using IEEE 802.11 Technology , 2006, 2006 1st Workshop on Operator-Assisted (Wireless Mesh) Community Networks.

[2]  Odlyzko Andrew Data Networks are Lightly Utilized, and Will Stay That Way , 1999 .

[3]  Tzi-cker Chiueh,et al.  Experiences in building a multihoming load balancing system , 2004, IEEE INFOCOM 2004.

[4]  Raghupathy Sivakumar,et al.  A Transport Layer Approach for Achieving Aggregate Bandwidths on Multi-Homed Mobile Hosts , 2005, Wirel. Networks.

[5]  Donald F. Towsley,et al.  Exploring the Performance Benefits of End-to-End Path Switching , 2004, ICNP.

[6]  Jon Crowcroft,et al.  Architecting Citywide Ubiquitous Wi-Fi Access , 2007, HotNets.

[7]  Henning Schulzrinne,et al.  Connection Sharing in an Ad Hoc Wireless Network among Collaborating Hosts , 1999 .

[8]  Saleem N. Bhatti,et al.  Coalition-Based Peering for Flexible Connectivity , 2006, 2006 IEEE 17th International Symposium on Personal, Indoor and Mobile Radio Communications.

[9]  Anees Shaikh,et al.  Multihoming Performance Benefits: An Experimental Evaluation of Practical Enterprise Strategies , 2004, USENIX Annual Technical Conference, General Track.

[10]  Paramvir Bahl,et al.  MultiNet: connecting to multiple IEEE 802.11 networks using a single wireless card , 2004, IEEE INFOCOM 2004.

[11]  Hari Balakrishnan,et al.  A measurement study of vehicular internet access using in situ Wi-Fi networks , 2006, MobiCom '06.

[12]  Marcelo Bagnulo,et al.  Shim6: Level 3 Multihoming Shim Protocol for IPv6 , 2009, RFC.

[13]  Paul Barford,et al.  Generating representative Web workloads for network and server performance evaluation , 1998, SIGMETRICS '98/PERFORMANCE '98.

[14]  Pablo Rodriguez,et al.  MAR: a commuter router infrastructure for the mobile Internet , 2004, MobiSys '04.

[15]  Haiyun Luo,et al.  Flow Scheduling for End-Host Multihoming , 2006, Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications.

[16]  Nicolas Christin,et al.  Taking Advantage of Multihoming with Session Layer Striping , 2006, Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications.

[17]  Yin Zhang,et al.  Optimizing cost and performance for multihoming , 2004, SIGCOMM '04.

[18]  Randall Stewart,et al.  SCTP: New Transport Protocol for TCP/IP , 2001, IEEE Internet Comput..