A stop-and-go queueing framework for congestion management
SIGCOMM '90 Proceedings of the ACM symposium on Communications architectures & protocols
ACM SIGCOMM Computer Communication Review
vic: a flexible framework for packet video
Proceedings of the third ACM international conference on Multimedia
A flexible model for resource management in virtual private networks
Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
| Hi-index | 0.00 |
Circuit networks are expensive to build, difficult to operate, fragile, and not easily scalable. Many network operators would like to carry circuit traffic as an overlay on top of an IP network. With higher bandwidths, faster routers, and Differentiated Services [RFC2474, RFC2475] features in routers, this is now possible. In this paper we describe a simple set of mechanisms that are sufficient to allow an IP cloud to carry circuit replacement traffic. Then, using the framework of [RFC3086] for describing a Per-Domain Behavior (PDB), we explain where these mechanisms should be deployed and quantify how they should be configured in order to construct the appropriate edge-to-edge behavior. This "Virtual Wire" (VW) PDB makes it possible to replace dedicated circuits with IP transport. In the process we attempt to lay to rest two pieces of QoS mythology: first that a Diffserv approach requires substantial over-provisioning compared to int-serv, and second that Diffserv quality of service is inherently loose and not quantitative.