In the price-based allocation scheme, the resource should be allocated fairly among users, and the system must reach a steady state. Users access resource according to the price in time slots. But when users determine their demand at the start of time slot, the resource price of that slot isn't yet known, it needs to be estimated. After the aggregate demand and the real price are computed, the price estimation will be updated to adapt to the change in demand. The iterative tatonnement process is often used to update the estimation. But using this algorithm, the allocation scheme may not converge. This paper proposes a novel method to determine the demand by considering the relation between the price changes with the aggregate demand change. Doing so, the convergence and the fairness are achieved. The conclusion is verified by simulation result.
Published in:
IT in Medicine & Education, 2009. ITIME '09. IEEE International Symposium on
(Volume:1
)
Date of Conference: 14-16 Aug. 2009