Flow Rate Fairness: Dismantling a Religion

Network engineer Richard Bennett’s new article for The Register: Dismantling a Religion: The EFF’s Faith-Based Internet explores the difference between the way the EFF wants to see the the Internet managed and current discussions under way in the IETF.

Bottom line: the Internet has never had a user-based fairness system, and it needs one. All networks need one.

EFF’s contention is that the freedom to connect is sacred. While bandwidth throttling or other controls of that nature are ok, Comcast’s current policy of ‘forging’ resets to BitTorrent clients is evil in principle.

Bennett goes into the nuts and bolts of the DOCSIS system used on cable networks to show that the actual number of packets transmitted is as crucial to network performance as the bandwidth used. Cutting the number of connections that a BitTorrent client uses can actually improve throughput without impeding access.

The internet’s traditional method of ensuring fairness – dropping packets – doesn’t work any more for any network that hosts peer-to-peer file-sharing applications designed to grab all the bandwidth/connections they can get.

In March Bob Briscoe presented a paper to the IETF Flow Rate Fairness: Dismantling a Religion that attacked the problem head-on. Abstract:

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.

On his blog Bennett quotes a warning from TCP-Friendly:

The network will soon begin to require applications to perform congestion control, and those applications which do not perform congestion control will be harshly penalized by the network (probably in the form of preferentially dropping their packets during times of congestion).

About joly

isoc member since 1995

Leave a Reply