Workload Generation and Instrumentation for Reproducible Networking Experiments
The authors propose the design of a benchmarking and instrumentation component for the GENI experimental networking test-bed. Their design uses a workload generation system to create reproducible network conditions and an instrumentation component to collect measurement data. They discuss the research challenges related to this design. Research on the design and implementation of computer networks requires experimental validation. As with any complex engineered system, theoretical models and simulation are useful in the early stages of the design process, but experimentation is necessary to obtain results on actual system performance under realistic conditions.