| Network research generally requires a simulation or emulation environment to test and evaluate the performance of protocols, algorithms, services, and applications, or to study the complex and highly varying network behaviors in both wireline networks and wireless networks. When the target network is sufficiently large, simulation of the target network will consume a large amount of time and memory, and its result is mostly based on many modeling assumptions. Existing network emulators can only support end-to-end network emulation because they fail to emulate network topology. In addition, network emulators are non-scalable due to the limitation of available physical infrastructure and the one-to-one mapping scheme. In this research, we present a scalable distributed network emulator cluster named EMPOWER, which not only can be used to emulate a large network with moderate cost, but also can generate user-defined network conditions and traffic dynamics at packet level. EMPOWER is highly scalable in that each emulator node could be configured to emulate multiple network nodes according to some predefined topology. By configuring the emulated network nodes on each emulator node across the entire emulator cluster, an emulated large scale network topology can be created using a considerably limited number of emulator nodes, making it possible to facilitate topology related research. Some significant research issues such as network topology mapping, resource competition in an emulator node, the overhead of emulation, and wireless network emulation, are discussed and addressed. Our emulation experiment with the prototype show that EMPOWER is capable of assisting the study of both wireline and wireless network protocols and applications. |
