Congestion allocation for distributed networks: an experimental study
- 85 Downloads
This paper reports an experimental study of two prominent congestion and cost allocation mechanisms for distributed networks. We study the fair queuing (or serial) and the FIFO (or average cost pricing) mechanisms under two different treatments: a complete information treatment and a limited information treatment designed to simulate distributed networks. Experimental results show that the fair queuing mechanism performs significantly better than FIFO in all treatments in terms of efficiency, predictability measured as frequency of equilibrium play, and the speed of convergence to equilibrium.
KeywordsSerial cost sharing Congestion allocation Experiment
JEL Classification NumbersC91 D83
Unable to display preview. Download preview PDF.
- Demers A., Keshav S., Shenker S. (1990) Analysis and simulation of a fair queueing algorithm. J Internetwork 1, 3–26Google Scholar
- Fischbacher U.: z-tree: A toolbox for readymade economic experiments—experimenter’s manual. University of Zurich Working Paper No. 21 (1999)Google Scholar
- Friedman E., Shenker S.: Learning and implementation on the internet. Manuscript: Rutgers University, Department of Economics (1998)Google Scholar
- Huck S., Sarin R.: Players with limited memory. Contrib Theoret Econ 4, Article 6 (2004)Google Scholar
- McKelvey R.D., McLennan A.M., Turocy T.L.: Gambit: software tools for game theory, version 0.2006.01.20. 2006. http://econweb.tamu.edu/gambitGoogle Scholar
- Razzolini L., Reksulak M., Dorsey R.: An experimental evaluation of the serial cost sharing rule. Manuscript: University of Mississippi (2007)Google Scholar
- Shenker S. Making greed work in networks: a game-theoretic analysis of gateway service disciplines. Sigmetrics 241–242 (1990)Google Scholar
- Stoica I., Shenker S., Zhang H.: Core-stateless fair queueing: Achieving approximately fair bandwidth allocations in high speed networks. Proceedings of ACM SIGCOMM, 118–130 (1998)Google Scholar