Thabet, Inès, Hanachi, Chihab and Ghédira, Khaled (2013) A Framework for an adaptive grid scheduling: an organizational perspective. In: Transactions on Computational Collective Intelligence Springer Berlin Heidelberg. Series “Lecture Notes in Computer Science”, Vol. 8065. pp. 52-87.

[thumbnail of assistant_3462520_1633667815_0.pdf]
Preview
Text
Download (10MB) | Preview

Abstract

Grid systems are complex computational organizations made of several interacting components evolving in an unpredictable and dynamic environment. In such context, scheduling is a key component and should be adaptive to face the numerous disturbances of the grid while guaranteeing its robustness and efficiency. In this context, much work remains at low-level focusing on the scheduling component taken individually. However, thinking the scheduling adaptiveness at a macro level with an organizational view, through its interactions with the other components, is also important. Following this view, in this paper we model a grid system as an agent-based organization and scheduling as a cooperative activity. Indeed, agent technology provides high level organizational concepts (groups, roles, commitments, interaction protocols) to structure, coordinate and ease the adaptation of distributed systems efficiently. More precisely, we make the following contributions. We provide a grid conceptual model that identifies the concepts and entities involved in the cooperative scheduling activity. This model is then used to define a typology of adaptation including perturbing events and actions to undertake in order to adapt. Then, we provide an organizational model, based on the Agent Group Role (AGR) meta-model of Freber, to support an adaptive scheduling at the organizational level. Finally, a simulator and an experimental evaluation have been realized to demonstrate the feasibility of our approach.

Item Type: Book Section
Language: English
Date: 2013
Uncontrolled Keywords: Grid scheduling - Multi-agent system - Adaptation
Subjects: H- INFORMATIQUE
Divisions: Institut de Recherche en Informatique de Toulouse
Site: UT1
Date Deposited: 20 Mar 2019 11:20
Last Modified: 02 Apr 2021 15:59
URI: https://publications.ut-capitole.fr/id/eprint/30607
View Item

Downloads

Downloads per month over past year