Information Technology Reference
In-Depth Information
through a dynamic pricing strategy. Various multi-access and multi-functional auction
systems were designed and implemented, including e-Bay, Allegro or Merkato.
Several researches had tackled with the problem of decision making based on a
specific auction policy named Vickrey auction . The basic idea of this kind of auction
algorithms is that the price, which should be paid by the winners, is lower than de-
clared [1], [6], [7], [10], [11].
In this paper two new auction-based scheduling algorithms were introduced and
Vickrey mechanism for pricing offers was applied. The paper considers a combined
approach of auction algorithm, QoS and admission control of requests, related to
individual clients. The main contribution of this paper is an application of new
mechanism of auction based on admission control of user requests arriving to the web
service. Our proposal is featured with a charging schema in respect with acceptable
QoS levels and closely connected with a client satisfaction. Proposed algorithms
guarantee the time delivering of service resources (i.e. multimedia files, audio topics
files, topics as text file) [10].
The process of offers collection from web clients will be discussed in the second
section. The third section formulates the problem. In the fourth section the auction
based scheduling mechanisms will be proposed. The fifth section presents simulation
results. In the last section the conclusions are given.
2 User Offers
The role of web service will be to schedule web client requests and to guarantee them
defined service level (defined bandwidth) for downloading files. Every user (web cli-
ent) will be obtained to reveal his offer. Every offer ( o n = {r j ,p u i k , b u i k } n ) will be char-
acterized by triple of following variables: resource name, price, and bandwidth.
Let's analyze an example: offer o 1 =[r 1 , 25$, 128 kb/s] includes resource r 1 for 25$
for bandwidth 128 kb/s, offer o 2 =[r 1 , 40$, 192 kb/s] includes resource r 1 for 40$ for
bandwidth 192 kb/s. Allowable is to report many offers for one resource by one cli-
ent, but the declared price for increasing number of bandwidth pieces should reflect
non-decreasing valuation function.
3 Problem Formulation
The aim of the web service will be to manage of the bandwidth pieces assuring as-
signed transmission speed. The service will support delivering of data files, for which
web client will be obligated to pay a charge related to requested bandwidth.
The problem will be formulated as follows:
For given:
- o 1 , …., o N - the set of offers, each offer contains the resource name, price and auc-
tioned bandwidth,
- R - the set of data files (resources), r j є R ,
- T - scheduling time interval,
- B - total available bandwidth,
Search WWH ::




Custom Search