GALO: A DEPLOYABLE FRAMEWORK FOR PROVIDING BETTER THAN BEST-EFFORT QUALITY OF SERVICE

Abstract

In this paper we propose a deployable approach to improving QoS by using a generic, extendable, overlay architecture; the Generalized Application Layer Overlay (GALO). The goals of this work are to 1) create an overlay architecture which allows us to sample specific path quality metrics among different paths; 2) utilize the proposed overlay architecture in order to implement our proposed QoS-based routing schemes, Application Layer Switching (ALSW) and Application Layer Striping (ALST). Perhaps the most significant contribution of this research is that we are able to achieve better than best-effort QoS without modifying intermediate nodes (i.e., routers), thus encouraging immediate deployment. Additionally, this research is performed on an actual wide area network (WAN) testbed, comprised of universities across the nation. Also, we assemble this architecture as a peer-to-peer framework, encouraging collaborating individuals with average workstations to improve the QoS of their traffic.