The networking research community lacks a tradition of sharing experimental data, or using such data for reproducing results. But are we really that bad? Are we worse than researchers in other fields?

Anomalous events that affect the performance of networks are a fact of life. It is therefore not surprising that recent years have seen an explosion in research on network anomaly detection. What is quite surprising, however, is the lack of controlled evaluation of these detectors. In this paper we argue that there are numerous important questions regarding the effectiveness of anomaly detectors that cannot be answered by the evaluation techniques employed today. We present four central requirements of a rigorous evaluation that can only be met by simulating both the anomaly and its surrounding environment. While simulation is necessary, it is not sufficient. We therefore present an outline of an evaluation methodology that leverages both simulation and traces from operational networks.

In this paper we propose a radical solution to data hosting and delivery for the Internet of the future. The current data delivery architecture is “network centric”, with content stored in data centers connected directly to Internet backbones. This approach has multiple drawbacks among which complexity of deploying data centers, power consumption, and lack of scalability are the most critical ones. We propose a totally innovative and orthogonal approach to traditional data centers, through what we call “nano” data centers, which are essentially boxes deployed at the edge of the network (e.g., in home gateways, set-top-boxes, etc.) that cooperate in a peer-to-peer manner. Unlike traditional peer-to-peer clients, however, our nano data centers operate under a common management authority, e.g., the ISP who installs and maintains the set-top-boxes, and can thus cooperate more effectively and achieve a higher aggregate performance. Nano data centers are, therefore, better suited for providing guaranteed quality to new emerging applications such as online gaming, interactive IPTV and VoD, and user generated content.

The available bandwidth of a path directly impacts the performance of throughput sensitive applications, e.g., p2p content replication or podcasting. Several tools have been devised to estimate the available bandwidth. The vast majority of these tools follow either the Probe Rate Model (PRM) or the Probe Gap Model (PGM).

Despite the flurry of anomaly-detection papers in recent years, effective ways to validate and compare proposed solutions have remained elusive. We argue that evaluating anomaly detectors on manually labeled traces is both important and unavoidable. In particular, it is important to evaluate detectors on traces from operational networks because it is in this setting that the detectors must ultimately succeed. In addition, manual labeling of such traces is unavoidable because new anomalies will be identified and characterized from manual inspection long before there are realistic models for them. It is well known, however, that manual labeling is slow and error-prone. In order to mitigate these challenges, we presentWebClass, a web-based infrastructure that adds rigor to the manual labeling process. WebClass allows researchers to share, inspect, and label traffic timeseries through a common graphical user interface. We are releasing WebClass to the research community in the hope that it will foster greater collaboration in creating labeled traces and that the traces will be of higher quality because the entire community has access to all the information that led to a given label.

As significant resources are directed towards clean-slate networking research, it is imperative to understand how cleanslate architectural research compares to the diametrically opposite paradigm of evolutionary research. This paper approaches the “evolution versus clean-slate” debate through a biological metaphor. We argue that evolutionary research can lead to less costly (more competitive) and more robust designs than clean-slate architectural research. We also argue that the Internet architecture is not ossified, as recently claimed, but that its core protocols play the role of “evolutionary kernels”, meaning that they are conserved so that complexity and diversity can emerge at the lower and higher layers. We then discuss the factors that determine the deployment of new architectures or protocols, and argue, based on the notion of “auto-catalytic sets”, that successful innovations are those that become synergistic components in closed loops of existing modules. The paper closes emphasizing the role of evolutionary Internet research.