Fig 4 - uploaded by Attila Kertész
Content may be subject to copyright.
Source publication
The chapter summarizes activities of COST IC1304 ACROSS European Project corresponding to traffic management for Cloud Federation (CF). In particular, we provide a survey of CF architectures and standardization activities. We present comprehensive multi-level model for traffic management in CF that consists of five levels: Level 5 - Strategies for...
Context in source publication
Context 1
... no. 1 (see Fig. 3): this is the reference scheme when the clouds work alone, denoted by SC. -Scheme no. 2 (see Fig. 4): this scheme is named as full federation and assumes that all clouds dedicate all theirs resources and clients to the CF system. This scheme we denote as FC. -Scheme no. 3 (see Fig. 5): for this scheme we assume that each cloud can delegate to CF only a part of its resources as well as a part of service requests coming from its ...
Citations
... 5. Interoperability It is indispensable to guarantee communication and data interchange between different clouds. 6. Monitoring It has two types. ...
... Instead, the federation management research category addresses several aspects of resource management [3], traffic management [6], and energy management [42], etc. ...
In order to provide uninterrupted services and fulfill the requirements of customers, Horizontal Cloud Federation Formation (HCFF) has been proposed as a new model enabling cloud providers to cooperate and enlarge their virtual machine infrastructure capacity. In this paper, we have established a synthesis on the main related works in the literature by classifying them into two classes: reactive protocols of HCFF and proactive protocols of HCFF, while doing a comparison between them according to a number of well chosen criteria. Furthermore, we have proposed three protocols of cloud federation formation based on the theory of cooperative games. The protocols are: Sequential Protocol of Horizontal Cloud Federation Formation (SPHCFF), Parallel PHCFF (PPHCFF), and Overlapping PHCFF (OPHCFF). More precisely, our main contribution in this research is the introduction of the notion “Overlapping Federations” with the OPHCFF protocol by referring to the class of Overlapping Coalition Formation Games. Moreover, we have developed a new system of three components based on Inter-Cloud architecture, and then we have implemented it in real time to evaluate the proposed protocols. The obtained experimental results have shown that the protocol OPHCFF improves the average response time, the rate of processed applications, and the profit compared with the two other introduced protocols SPHCFF and PPHCFF. In addition, despite the dissimilarities of architectures, protocols, and mathematical models of existing works, we have arrived to manage a simulation comparison of our SPHCFF with a concurrent mechanism called Cloud Federation Formation Mechanism (CFFM). The obtained simulation results have shown that the proposed SPHCFF performs better than the CFFM mechanism, when the overload rate is large enough, in terms of profit, federation size, time execution, and failure rate.
... Distributed Cloud Computing has definitely changed the scene of data/information technology by giving some real advantages to IT clients, including dispensing with forthright IT venture, scalability, relative expenses and so on (Ghahramani, 2017;Zheng, 2017). In Wojciech Burakowski (2018), the idea of distributed cloud computing frameworks reaching out to Cloud Federation (CF) by combining various clouds into one framework is presented. Cloud service providers work in geologically distributed fashion where different servers take on client requests like calamity recuperation and multi-site reinforcements which ended up across the board. ...
Fog computing is an encouraging computational model that extends distributed cloud computing to the edge of systems. It varies to cloud computing with some of the attributes. Fog computing has new challenges while building and maintaining the trust among the fog nodes and with edge devices. The solutions applied for the various cloud challenges cannot be directly applied for fog computing. This chapter gives an overview of these difficulties and relates solutions in a concise way. It also highlights the open challenges that still exist in fog computing.
