Please use this identifier to cite or link to this item:
Title: Distributed cloud services based on programmable agile networks
Authors: Filiposka, Sonja 
Demchenko, Yuri
Karaliotas, Tasos
de Vos, Migiel
Regvart, Damir
Keywords: Agile programmable network infrastructure, marketplace, multi-cloud services, service function chaining, zero touch provisioning
Issue Date: 13-Jun-2016
Journal: Proceedings of TERENA networking conference (TNC16), Prague
Abstract: In order to allow researchers to focus on their work, all newly designed applications and services must not only offer features needed to tackle Big Data problems, but need to work seamlessly and intuitively, while efficiently hiding all non-necessary technical and networking details from the end user. Thus, today’s requirements of the research and education community demand a holistic converged approach in the design of new generation intelligent cloud aware networks that need to work in concert with the distributed application components. The goal of this paper is to contribute toward setting up the building blocks of an open multi-cloud ecosystem based on a programmable cloud service delivery infrastructure. For these purposes we examine the possibility of introducing a third-party multi-cloud marketplace placed on top of a cloud aware agile network that can be reconfigured based on the defined multi-cloud application workflow. We discuss the overall architecture of the ecosystem and define the requirements for the underlying network infrastructure. The described proof of concept demonstrates the benefits of the proposed solution providing an example of how the concepts of network function virtualization and software defined networks can be leveraged in order to obtain an agile programmable network that will respond to the changes in the application traffic flow. With self-provisioning being acknowledged as a crucial component among service providers, we discuss all steps needed in order to translate user requests in a fully automated end-to-end service delivery and monitoring especially in a multi-domain environment.
Appears in Collections:Faculty of Computer Science and Engineering: Journal Articles

Files in This Item:
File SizeFormat 
08-paper-TNC2016-2.pdf417.14 kBAdobe PDFView/Open
Show full item record

Page view(s)

checked on Nov 26, 2022


checked on Nov 26, 2022

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.