Network Variation Experiments

Congestion control algorithms (CCAs) are challenged by a wide range of network conditions. For example, the bandwidth can vary over multiple orders of magnitude depending on which medium is used for transmissions. In network variation experiments, it is studied how network conditions affect the performance of CCAs.

Each experiment sets up a network scenario to evaluate a certain performance aspect of a CCA. For example, the shallow buffers experiment tests if a CCA remains efficient when the bottleneck queue has a small buffer size. A CCA should perform well across a wide range of network conditions.