Exploiting Parallel Networks Using Dynamic Channel Scheduling (Invited Paper) Lara B. Deek, Kevin C. Almeroth, Mike P. Wittie Khaled A. Harras Department of Computer Science Department of Computer Science UC Santa Barbara Carnegie Mellon University Santa Barbara, CA 93101 Qatar {laradeek, almeroth, mwittie}@cs.ucsb.edu kharras@cs.cmu.edu ABSTRACT Many researchers have been focusing on the outcomes and consequences of the rapid increase and proliferation of mo- bile wireless technologies. If it is not already the case, it will soon be rare for a user to be in a situation where absolutely no network connection exists. In fact, through numerous devices, users will soon expect to be connected in all places at all times. Through the great variety and increase in the capabilities of these devices, it is not a surprise to find a single user with many connection opportunities. As a re- sult, we believe that the next major evolution of wireless mobile networks will be in the exploitation of multiple net- work connections in parallel. Due to network heterogene- ity, the major challenge in such situations is to determine the way that these networks can be utilized to better serve different network applications. In this work, we propose a dynamic channel scheduling mechanism that adapts to the state of the available channels to provide more efficient us- age of network connectivity. We do so by observing channel throughput, creating a set of channel usage combinations, and then choosing the most efficient combination. We eval- uate an implementation of the proposed mechanism using emulation. Our results show that under realistic conditions our dynamic approach greatly improves cost delay metrics, and the overall user-perceived performance compared to a more static approach. Categories and Subject Descriptors C.2.1 [Computer-Communication Networks]: Store and forward networks; C.2.2 [Computer-Communication Net- works]: Applications; C.2.5 [Computer-Communication Networks]: Access schemes General Terms Performance, Experimentation Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. WICON ’08, November 17-19, 2008, Maui, Hawaii, USA. Copyright 2008 ICST 978-963-9799-36-3. 1. INTRODUCTION The increase in communication demands and the popu- larity of the Internet have prompted the development and expansion of wireless mobile technologies that provide con- nectivity in increasingly diverse environments and under highly variable conditions. Today’s users not only expect reliable network connectivity, but also demand the highest level of performance regardless of their location or under- lying network conditions. A promising way to meet user expectations and to provide greater user-satisfaction is to exploit the availability of multiple network connections in parallel. The ParaNets network architecture is an early ef- fort to use multiple heterogeneous networks in unison [1]. The system we propose in this paper builds on ParaNets to transparently unify available network connections into a single Internet service. Specifically in this paper, our goal is to provide the best possible performance subject to the cost of using individual networks and in response to varying network conditions. Heterogeneous parallel networks, such as cellular, satel- lite, and Wi-Fi, are characterized by different bandwidth, delay, and monetary cost. These characteristics make each of these networks best-suited for specific types of data trans- missions. For example, it may not be frugal to use high-cost satellite bandwidth for a movie download when a free Wi-Fi connection is available. Similarly, a user may not want to search for a free access point to send a short email when also already connected to a cellular data network. When multiple heterogeneous networks such as these are available, the major challenge is to determine the efficient use of each channel with respect to user performance expectations. The system we propose in this paper dynamically determines ef- ficient usage of available parallel networks based on traffic requirements as well as the state of these networks to provide a unified Internet service. We build our system over the ParaNets-Enabled Data Bundling System for Intermittent Connectivity (DBS-IC) architecture, originally developed to improve communica- tion over challenged networks [1]. ParaNets-Enabled DBS- IC takes advantage of the availability of multiple heteroge- neous networks in parallel to create a perception of constant connectivity in challenged networks. ParaNets-Enabled DBS- IC statically distributes data over multiple channels based on the expected performance of the channels’ underlying technologies, as well as traffic bandwidth and cost require- ments. While the static data allocation schemes achieve Digital Object Identifier: 10.4108/ICST.WICON2008.4969 http://dx.doi.org/10.4108/ICST.WICON2008.4969