Flow Rate Fairness: Dismantling a Religion

  • The service having id "facebook_widget" is missing, reactivate its module or save again the list of services.
  • The service having id "digg_smart_button" is missing, reactivate its module or save again the list of services.
Bob Briscoe
Appears in: 
CCR April 2007

Resource allocation and accountability keep reappearing on every list of requirements for the Internet architecture. The reason we never resolve these issues is a broken idea of what the problem is. The applied research and standards communities are using completely unrealistic and impractical fairness criteria. The resulting mechanisms don’t even allocate the right thing and they don’t allocate it between the right entities. We explain as bluntly as we can that thinking about fairness mechanisms like TCP in terms of sharing out flow rates has no intellectual heritage from any concept of fairness in philosophy or social science, or indeed real life. Comparing flow rates should never again be used for claims of fairness in production networks. Instead, we should judge fairness mechanisms on how they share out the ‘cost’ of each user’s actions on others.

Public Review By: 
Michalis Faloutsos

The paper addresses the related issues of fairness and resource accountability of Internet in its current form. Not being envisioned as a commercial product, Internet was designed with accountability near the bottom of its design priority list (see Clark’s paper in SIGCOMM’88). This paper does a great job at pointing out the many flaws of flow rate fairness, which although it is the predominant way of defining fairness, it does not seem to be practical. The paper follows an interesting approach between technical and commercial considerations. It also presents an alternative solution to the resource allocation problem. Some of the reviewers thought that the proposed scheme seemed promising but it may need to be developed more. In fact, the reviewers had very strong opinions about different issues regarding this paper, and it sparked an unusual amount of reiterations for a CCR manuscript. I consider this as a good sign, and I invite the community to express its opinions at CCR Online.