GridEcon: A Market Place for Computing Resources

Altmann, Jörn; Courcoubetis, Costas; Stamoulis, George D.; Dramitinos, Manos; Rayna, Thierry; Risch, Marcel; Bannink, Chris

doi:10.1007/978-3-540-85485-2_15

Cited by 52 publications

(44 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As explained in the introduction, the framework of our research are open Utility Computing Markets [4]. Such markets are dynamic: any small provider can spontaneously enter in the market and they must be automatically considered by the reputation system (for example, a company with few hosts for its internal use may sell its spare resources at night to help amortizing the cost of its infrastructure).…”

Section: Description Of the Reputation Modelmentioning

confidence: 99%

Analysis of a trust model for SLA negotiation and enforcement in cloud markets

Macías

Guitart

2016

Future Generation Computer Systems

View full text Add to dashboard Cite

Online Reputation Systems help mitigate the information asymmetry between clients and providers in Cloud Computing Markets. However, those systems raise two main drawbacks: the disagreement for assuming the cost of ownership of such services and their vulnerability to reputation attacks from dishonest parties that want to increase their reputation. This article faces both problems by describing a decentralised (peer-topeer) trust model that does not require the intervention of a central entity to manage it. This model includes mechanisms to allow participants to avoid dishonest behaviour from other peers: each client statistically analyses the external reports about providers and updates the trustworthiness of the peers. The trustworthiness values will be used to negotiate prices in later transactions. The trust model is then incorporated in the Service-Level Agreement negotiation and enforcement processes, prioritising trusted clients over non-trusted clients to minimise the consequences of low Quality of Service in relation to the trust of the provider, and incentivise accurate trust reports from the clients. Finally, this article evaluates and discusses the validity of the trust model under different attacks from dishonest clients and providers.

show abstract

Section: Description Of the Reputation Modelmentioning

confidence: 99%

Analysis of a trust model for SLA negotiation and enforcement in cloud markets

Macías

Guitart

2016

Future Generation Computer Systems

View full text Add to dashboard Cite

show abstract

“…One possible approach to sharing computational resources is to form an open market for computation, see [14], [16]. In this market providers (sellers) and consumers (buyers) of computing resources go to trade.…”

Section: Economic Issuesmentioning

confidence: 99%

Economic Issues in Shared Infrastructures

Courcoubetis¹,

Weber²

2012

IEEE/ACM Trans. Networking

Self Cite

View full text Add to dashboard Cite

We define some interesting incentive issues that arise in the management of virtual infrastructures. We demonstrate that participants' decisions about the quantities of infrastructure that they will choose to contribute to a virtual organization can be greatly affected by the resource sharing policy that they know will be deployed when the system operates. Unless this policy is well-designed, agents will attempt to free-ride by contributing less resource than is desirable. Our novel contribution is the formulation of models for designing optimal management policies, an analysis that demonstrates the inadequacy of simple sharing policies, and proposals for some better ones. We find an optimal policy in a limit as the number of participants becomes large. We learn that simple policies may be far from optimal and that efficient policy design is not trivial; policy parameters play important role in optimizing the efficiency of virtual facility formation.

show abstract

“…Risch et al [7] designed a cloud services testbed to test cloud mechanisms. Altmann et al [8] designed an exchange market for grid resources, which was tested on the platform designed by Risch et al [7]. Such a marketplace for clouds is not easy to build because of interoperability issues, therefore the current focus on the cloud markets is mostly on the single provider and multiple users model.…”

Section: Introductionmentioning

confidence: 99%

An Online Mechanism for Dynamic VM Provisioning and Allocation in Clouds

Zaman

Grosu

2012

2012 IEEE Fifth International Conference on Cloud Computing

View full text Add to dashboard Cite

Abstract-Current cloud computing providers allocate their virtual machine (VM) instances via fixed price-based or auctionlike mechanisms. However, these mechanisms have one limitation, they are all offline mechanisms, therefore they need to collect information and be invoked periodically. In this paper, we address this limitation by designing an online mechanism for dynamic provisioning and allocation of VM instances in clouds. Our proposed mechanism, MOVMPA, is invoked as soon as a user places a request or some VM instances already allocated become available again. When invoked, the mechanism selects users who would be allocated VM instances for the period they requested for, and ensures that those users will continue using those VMs for the entire period requested. We prove that the mechanism is incentive compatible and also investigate its performance through extensive simulation experiments.

show abstract

GridEcon: A Market Place for Computing Resources

Cited by 52 publications

References 3 publications

Analysis of a trust model for SLA negotiation and enforcement in cloud markets

Analysis of a trust model for SLA negotiation and enforcement in cloud markets

Economic Issues in Shared Infrastructures

An Online Mechanism for Dynamic VM Provisioning and Allocation in Clouds

Contact Info

Product

Resources

About