Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/95966
Title: Grid resource allocation by means of option contracts
Authors: Bossenbroek, Anton
Tirado-Ramos, Alfredo
Sloot, Peter M. A.
Keywords: DRNTU::Engineering::Computer science and engineering
Issue Date: 2009
Source: Bossenbroek, A., Tirado-Ramos, A., & Sloot, P.M.A. (2009). Grid Resource Allocation by Means of Option Contracts. IEEE Systems Journal, 3(1), 49-64.
Series/Report no.: IEEE systems journal
Abstract: In Grid environments, where virtual organization resources are allocated to users using mechanisms analogue to market economies, strong price fluctuations can have an impact on the nontrivial quality-of-service expected by end users. In this paper, we investigate the effects of the use of option contracts on the quality of service offered by a broker-based Grid resource allocation model. Option contracts offer users the possibility to buy or sell Grid resources in the future for a strike price specified in a contract. By buying, borrowing and selling option contracts using a hedge strategy users can benefit from expected price changes. In this paper, we consider three hedge strategies: the butterfly spread which profits from small changes, the straddle which benefits from large price changes, and the call strategy which benefits from soaring prices. Using our model based on an abstract Grid architecture, we find that the use of hedge strategies augment the ratio of successfully finished jobs to failed jobs. We show that the degree of successfulness from hedge strategies changes when the number of contributed resources changes. By means of a model, we also show that the effects of the butterfly spread is mainly explained by the amount of contributed resources. The dynamics of the two other hedge strategies are best explained by observing the price behavior. We also find that by using hedge strategies the users can increase the probability that a job will finish before the deadline. We conclude that hedging using options is a promising approach to improve resource allocation in environments where resources are allocated by using a commodity market mechanism.
URI: https://hdl.handle.net/10356/95966
http://hdl.handle.net/10220/10175
DOI: http://dx.doi.org/10.1109/JSYST.2008.2011255
Rights: © 2009 IEEE.
Fulltext Permission: none
Fulltext Availability: No Fulltext
Appears in Collections:SCSE Journal Articles

Google ScholarTM

Check

Altmetric

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.