On the impact of network QoS on automated distributed auctions

Autonomic networks offer a promising technology for the development of future distributed services, which will benefit from self-adapting communications with consequent improvements in quality of service (QoS). As electronic commerce over shared computer networks continues to grow, it becomes meaningful to gain better understanding of the relationship between network QoS and the expected results of time-sensitive e-commerce applications. In particular, to quantify possible utility improvements with the introduction of autonomic networks. This work is a simulation study of a time critical e-commerce application in the form of a system of distributed auctions that relies on a computer network for marketing, matching sellers and buyers, and conducting automated auctions and transactions. Simulation results show important reductions in utility for sellers under degrading QoS conditions, therefore suggesting the marginal gains of autonomic network support. The results also suggest that observations of the network should be considered in the design of effective trading strategies for such systems.