Large-scale virtualization in the Emulab network testbed
ATC'08 USENIX 2008 Annual Technical Conference on Annual Technical Conference
Interface connecting the INET simulation framework with the real world
Proceedings of the 1st international conference on Simulation tools and techniques for communications, networks and systems & workshops
A real-time network simulation infrastructure based on OpenVPN
Journal of Systems and Software
Hybrid testbeds for QoS management in opaque MANETS
Proceedings of the 4th Annual International Conference on Wireless Internet
A Non-intrusive Estimation for High-Quality Internet TV Services
FMN '09 Proceedings of the 2nd International Workshop on Future Multimedia Networking
VP2P: a virtual machine-based P2P testbed for VoD delivery
CCNC'09 Proceedings of the 6th IEEE Conference on Consumer Communications and Networking Conference
A virtual ad hoc network testbed
International Journal of Communication Networks and Distributed Systems
Integrated network experimentation for QoS measurements in opaque MANETs
International Journal of Network Management
EURASIP Journal on Wireless Communications and Networking - Special issue on simulators and experimental testbeds design and development for wireless networks
A non-intrusive estimation for high-quality Internet TV services
Multimedia Tools and Applications
Hi-index | 0.00 |
Discrete-event packet-level network simulation is well-known and widely used. Network emulation is a hybrid approach that combines real elements of a deployed networked application-such as end hosts and protocol implementations-with synthetic, simulated, or abstracted elements-such as the network links, intermediate nodes and background traffic. A key difference between the two approaches is that in the former, the notion of time is virtual and is independent of real time, whereas the latter must execute in real time. Emulation gains realism while naturally foregoing complete repeatability; historically, emulation was also tedious to control and manage. We define integrated network experimentation as spatially combining real elements with simulated elements in the same experimental run, each modeling different portions of anetwork topology. Integrated experiments enable new validation techniques and larger experiments than obtainable by using real elements alone. This paper highlights the key issues in integrated network experimentation, and presents some of the design techniques we use in designing, building, and putting into public production use such an integrated environment, running on a space-shared cluster.