With the Internet becoming ubiquitous and a critical infrastructure in the service economy, the demand to capture traffic or to document relevant resource usage characteristics arises for manifold reasons. Network operators, policy makers, and end-users alike are interested in accounting data, albeit not always for similar motives. The type and granularity of data to be collected depends on these diverging demands wherefore different accounting techniques allowing to extract distinct properties at certain granularities become necessary.
As in many cases not all traffic, but only certain traffic characteristics are of interest, capturing of each entire package is often not necessary. However, this focal point in characteristics makes a capturing in central links of a network most valuable. Unfortunately, these central links nowadays often provide data rates at which accounting at a given granularity becomes either expensive or infeasible.
Beyond purely technical difficulties, the investigation of high-speed accounting is interesting, since it concerns stakeholders from many different areas. Therefore, high-speed accouting shows a strong socio-economic component which may raise regulatory concerns. For example, high-speed accounting systems facilitate the collection and persistent storage of a vast amount of (raw or filtered) data — data that may be classified as personal data documenting end user behavior, allowing possibly mass profiling. Accordingly, high-speed accounting is confronted with privacy concerns for the possibility of excessive data gathering. While Internet Service Providers (ISP) are, in general, assumed to deploy high-speed accounting primarily for economic or managerial reasons, they have to find consensus with regulators (e.g., data protection offices) in order to meet the relevant set of privacy requirements.
Another example for regulatory concerns in relation to high-speed accounting may be the case of an ISP, that evaluates adopting volume-based charging: The motivation for an ISP to offer volume-based tariffs arises in the managerial field. Volume-based tariffs would allow an ISP to give incentives to its customers to not overuse network resources, e.g., by heavily sharing data. A more fair charging would be in place, according to the measure of relative usage of a shared medium. Since Internet congestion is a serious threat to operability, ISPs see a clear managerial motivation to offer volume-based charging. However, volume-based charging requires per definition a working and fine-granular accounting being in place in order to produce the data basis based on which volumes may be calculated. In high-speed links, the question arises though if collecting such data is technically feasible at all. Furthermore, out of economic considerations, the question arises, if the revenue gained by volume-based charging outweighs the cost of the necessary accounting infrastructure. What is more, is the legally driven question whether the temporal and content-wise granularity needed for a volume-based charging model is acceptable with respect to data privacy.
For the presented reasons, it is of large interest to evaluate the overall deployment of high-speed accounting with respect to technical feasibility as well as socio-economic and legal factors.
This assignment endorses three areas of work. The first addresses socio-economic aspects of high-speed accounting, where high-speed accounting is to be understood as accounting activities applied to links with data rates of 10 Gbit/s or higher. Different stakeholders that are affected by accounting and/or may have interest in it shall be identified and analyzed with respect to their respective incentives and range of options available. It is likely to find different or even conflicting interests of these stakeholders, i.e., different stakeholders may want to apply accounting in different ways whereat some of these ways are displeasing to other stakeholders. These conflicting interests may, in extreme cases, be decided by judicial interactions of stakeholders at odds.
The second area of work therefore addresses the determination of accounting activities that are currently legally feasible. The assignment shall, for instance, determine which traffic details are currently allowed to be measured and which must not be investigated by network operators. To this aim, relevant legal acts from within a primarily European range shall be identified, classified, and assessed. In different words, the assignment shall provide an overview of the European legal basis determining the regulatory limits of high-speed accounting.
Based on this collection and assessment of the legal basis in Europe, the third item of work shall shed light on the state of debate regarding (at least) three discussed topics in high-speed accounting. This item, hence, combines insights from the previous two work items in that it shall list and discuss stakeholders with their arguments for/against certain high-speed accounting practices — and it shall make explicit links to the investigated legal basis if possible. Furthermore, the analysis of debate shall also cover stakeholder arguments reflecting the use and/or limits of the current legal basis.