A decentralized self-adaptation mechanism for service-based applications in the cloud

Nallur, Vivek; Bahsoon, Rami

doi:10.1109/tse.2012.53

Cited by 88 publications

(61 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We also mention the decentralized self-adaptation mechanism in the cloud context using market-based heuristics [15]. It focused on the service selection problem at the application layer, while we are concerned with a planning problem at different layers in the cloud environment.…”

Section: Related Workmentioning

confidence: 99%

Decentralized Planning for Self-Adaptation in Multi-cloud Environment

Ismail

Cardellini

2015

Communications in Computer and Information Science

View full text Add to dashboard Cite

Abstract. The runtime management of Internet of Things (IoT) oriented applications deployed in multi-clouds is a complex issue due to the highly heterogeneous and dynamic execution environment. To effectively cope with such an environment, the cross-layer and multi-cloud effects should be taken into account and a decentralized self-adaptation is a promising solution to maintain and evolve the applications for quality assurance. An important issue to be tackled towards realizing this solution is the uncertainty effect of the adaptation, which may cause negative impact to the other layers or even clouds. In this paper, we tackle such an issue from the planning perspective, since an inappropriate planning strategy can fail the adaptation outcome. Therefore, we present an architectural model for decentralized self-adaptation to support the crosslayer and multi-cloud environment. We also propose a planning model and method to enable the decentralized decision making. The planning is formulated as a Reinforcement Learning problem and solved using the Q-learning algorithm. Through simulation experiments, we conduct a study to assess the effectiveness and sensitivity of the proposed planning approach. The results show that our approach can potentially reduce the negative impact on the cross-layer and multi-cloud environment.

show abstract

Section: Related Workmentioning

confidence: 99%

Decentralized Planning for Self-Adaptation in Multi-cloud Environment

Ismail

Cardellini

2015

Communications in Computer and Information Science

View full text Add to dashboard Cite

show abstract

“…However, some of these models assume that all the agents are acquaintance of each other (fully connected network) that is not a realistic situation in open environments and also rely on the hierarchical structure that reduces the flexibility of the system. There are other approaches that consider self-organization mechanisms in order to provide a service composition that deals with an specific goal [Khondoker et al 2011;Nallur and Bahsoon 2012]. The main drawback of these approaches is that they assumed that there is a global view of all the services available in the system and that broadcast mechanisms are used.…”

Section: Emergence and Self-organizationmentioning

confidence: 99%

Utility-Based Mechanism for Structural Self-Organization in Service-Oriented MAS

Val

Rebollo

Vasirani

et al. 2014

ACM Trans. Auton. Adapt. Syst.

View full text Add to dashboard Cite

The structural relations established among agents has influence on the performance of decentralized service discovery process in Multi-Agent Systems (MAS). Moreover, distributed systems should be able to adapt their structural relations to changes in environment conditions. In this paper, we present a Service-Oriented MAS where agents initially self-organize their structural relations based on the similarity of their services. Agents integrate a mechanism that during the service discovery process facilitates the self-organization of their structural relations in order to adapt the structure of the system to the service demand. This mechanism facilitates the task of decentralized service discovery and improves its performance. Each agent has local knowledge about its direct neighbors and the queries received during discovery processes. With this information, an agent is able to analyze its structural relations and decide when it is more appropriate to modify its direct neighbors and select the most suitable acquaintances to replace them. The experimental evaluation shows how this self-organization mechanism improves the overall performance of the service discovery process in the system when service demand changes.

show abstract

“…Decentralized mechanism has been studied widely in different applications based on intelligent agents [8]. The notion behind any multi-agent system is to divide a complex problem handled by a single entry (a centralized system) into smaller simpler problems to be handled by various entities as in a distributed system.…”

Section: Introductionmentioning

confidence: 99%

Decentralized Traffic Control Using Agents in Ethernet Passive Optical Networks (EPON)

Sadon¹,

Hizan²,

Kanesan³

et al. 2016

View full text Add to dashboard Cite

Traditionally, allocating of Ethernet Passive Optical Network (EPON) bandwidth schemes relied on a centralized architecture where their Optical Line Terminal (OLT) is the sole intelligent device with the capability for the arbitration of bandwidth based on time-division access for the upstream shared channel. However, any breakdown in the OLT bandwidth allocation will affect the allotment for Optical Network Units (ONUs). Few researches had dealt with a decentralized approach for the EPON and most of the solutions proposed involved additional cost by adding new complex devices to the original architecture. This paper proposes an intelligent decentralized mechanism in EPON that enhances the network performance using intelligent agents specification based on the IEEE 802.3 ah standards in the foundation for intelligent physical agent which is compatible with an Internet Protocol (IP)-based network. Specifically, this paper proposes a novel distributed scheme that supports differentiated services and ensures QoS for both Inter-and Intra-ONU allocation. The proposed mechanism introduces a unique direct communication between the ONUs supported by identical dynamic bandwidth allocation algorithms running simultaneously in each ONU. In addition to being fully distributed, the proposed scheme is asynchronous, scalable, and dynamic with flexibility and reliability in handling data, voice and video.

show abstract

A decentralized self-adaptation mechanism for service-based applications in the cloud

Cited by 88 publications

References 45 publications

Decentralized Planning for Self-Adaptation in Multi-cloud Environment

Decentralized Planning for Self-Adaptation in Multi-cloud Environment

Utility-Based Mechanism for Structural Self-Organization in Service-Oriented MAS

Decentralized Traffic Control Using Agents in Ethernet Passive Optical Networks (EPON)

Contact Info

Product

Resources

About