Virtualization and emulation have become worthy approaches to save significant amounts of money related to physical resource acquisition expenditures. In light of this, the development of network emulation platforms has led a revolution in the research and testing of novel services and applications, as they provide a cost-effective, flexible and reproducible environment for experimentation. However, they present some practical issues, in particular, their scalability is one of the limiting factors as it links the emulated networks that can be successfully deployed on a given hardware. We address this matter by testing the consumption and exploitation of physical resources of one popular network emulation platform, Mininet. We follow a methodology based on the isolation of the threads associated to the operating system, the virtual hosts, and the monitoring tasks. In such a manner, this approach can measure the effect of the placement of threads in the available cores, and help optimize bottlenecks that jeopardize the results of network emulations. Additionally, we monitor several key performance indicators for general-purpose Mininet deployments in different network topologies, varying the number of active elements, links and network conditions such as packet loss or delay. Our results show that Mininet presents performance bounds in commodity servers that suffice a wide range of general network tests. It achieves aggregated bandwidths above 10 Gb/s and median round-trip time values around 1 ms, even in demanding scenarios where more than a thousand hosts, up to 64-hop paths and 64 subnets are included in the emulated topologies.

Network managers have to deal with tons of measurement data provided by monitoring systems. Such data is difficult to both process and translate into concrete management actions. As an attempt to make managerial work easier, we propose a novel statistical approach that summarizes the behavior of network flow characteristics - e.g., flow sizes or durations. Bearing in mind that losses in the summarized information can lead to restricted or even erroneous conclusions, our approach solves this by exploiting the probability integral transform theorem. This theorem allows the definition of a set of intervals, mapped into concrete categories, where the number of flows according to a given characteristic would be uniformly distributed among categories. This eases the use of both statistical tests and simple visual inspection to detect changes in the behavior of the characteristic under analysis, as typically abrupt changes are understood as signs of intrusion, malfunction or other types of anomalies. This proposal gave rise to the visualization and analytical framework Dictyogram, which has been applied to monitor the Spanish Academic Network - more than one million users. Its results are shown as a case study assessing the usefulness of our proposal.