Market-Based Network Resource Allocation with Non-tâtonnement Process

Authors:
Kosuke Nakatsuka;Hirofumi Yamaki;Toru Ishida
Affiliations:
-;-;-
Venue:
PRIMA '00 Proceedings of the Third Pacific Rim International Workshop on Multi-Agents: Design and Applications of Intelligent Agents
Year:
2000

Citing 4
Cited 0

A Microeconomic Approach to Optimal Resource Allocation in Distributed Computer Systems

IEEE Transactions on Computers
Implementation Issues on Market-Based QoS Control

ICMAS '98 Proceedings of the 3rd International Conference on Multi Agent Systems
On Resource-Oriented Multi-Commodity Market Computations

ICMAS '98 Proceedings of the 3rd International Conference on Multi Agent Systems
A market-oriented programming environment and its application to distributed multicommodity flow problems

Journal of Artificial Intelligence Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

There are two types of market mechanism that can be used to allocate network resources efficiently based on users' preference. One, mainly considered in previous works, is the Tâtonnement process; the allocation is not changed until the price reaches equilibrium. When the user preference changes dynamically, the resource allocation derived by the Tâtonnement process may not reflect the curent preference. The other is the Non-Tâtonnement process, where the allocation is changed dynamically even while the allocation is being calculated. Consequently, it suits resource allocation in dynamic environments. However, the property of this process in terms of the time needed to complete all calculations has not been studied sufficiently. As a first step, this paper compares the quality of the results derived by the Non-Tâtonnement process to that by the Tâtonnement process for dynamic network resource allocation. We find that the Non-Tâtonnement process works better in rapidly changing environments and that the Tâtonnement process offers superior results when there is enough time for the market to converge to equilibrium.