An Instrumentation Framework for the Critical Task of Measurement Collection in the Future Internet
Experimentation for research on the future Internet involves observing multiple metrics at various distributed points of the network under observation. Be it in a testbed or in an uncontrolled field experiment, collecting these measurements is not an easy task. The relevant experimental data tends to be scattered across multiple hosts in different, incompatible formats, and sometimes buried amongst a trove of other measurements, irrelevant to the current question. Collecting, selecting and formatting the useful records is a time-consuming manual operation. In this paper, we present a conceptual framework to facilitate this task. It includes a common representation of any type of experimental data, as well as the elements to process and collect the measurement samples. We then introduce the OML suite, an instrumentation library following the concept of this framework. We outline its API and how it can be used to instrument an experiment in only a few lines of code. We also present an evaluation of this library, demonstrating that it can be used non-intrusively with respect to the processes and systems under observation, as well as confirming the validity of the proposed framework.
Keywords: Network measurements, Instrumentation system, OML, Performance evaluation