Poznań, Poland
Poznań, Poland

Time filter

Source Type

Toseef U.,EICT | Fernandez C.,I2CAT | Bermudo C.,I2CAT | Carrozzo G.,NXW | And 10 more authors.
Proceedings - European Workshop on Software Defined Networks, EWSDN | Year: 2015

The development of test environments as close as possible to the real world scenarios is becoming a fundamental requirement in designing innovative network applications. This environment must be fully configurable and reliable enough to provide similar results in multiple experiment runs. The federation of existing Future Internet (FI) test beds is an initiative tofu fill these strict requirements. The FELIX project aims to define, implement, and deploy a control and monitoring framework which allows experimenters to execute their network services in a distributed environment spread across two continents, i.e. Europe and Asia. This paper describes the architecture of the software components developed to manage heterogeneous resources that constitute the FELIX infrastructure, i.e. Computing, SDN and transport resources. This article introduces the components of a modular architecture with particular emphasis on the provided functionalities, the exported interfaces, the dependencies and the relationship between the internal building blocks. Details of the implementation choices and the workflows to realize user requests are also presented. © 2015 IEEE.

Landi G.,Nextworks | Tzanakaki A.,Athens Information Technology | Bernini G.,Nextworks | Parniewicz D.,PSNC | And 5 more authors.
2013 Future Network and Mobile Summit, FutureNetworkSummit 2013 | Year: 2013

New emerging high performance and distributed applications require the interconnection of IT resources through high capacity and flexible networks. In this context, the FP7 GEYSERS project proposes on-demand provisioning of converged Virtual Infrastructures composed of IT and optical network resources, able to support the delivery of cloud services. Virtual Infrastructures must be specifically designed to support the required service loads. However, due to the unpredictable volume and nature of services, dynamic and on-demand re-planning of Virtual Infrastructures is required to allow the operators to support service requests that are not known in advance, while limiting over-provisioning and optimizing the global infrastructure utilization. This paper investigates the business and technical motivations for infrastructure re-planning, and presents different re-planning models in terms of timescales, involved resources and procedure automation. A dedicated cross-layer architectural solution is proposed, describing its implementation within the GEYSERS prototype. Simulation studies are provided to compare the performance of static planning to that of dynamic re-planning, showing the re-planning benefits in terms of resource utilization and energy consumption. Finally, the paper presents the experimental validation of the prototype over the GEYSERS test-bed. © 2013 Authors.

Jofre J.,Distributed Applications and Networks Area | Velayos C.,Distributed Applications and Networks Area | Landi G.,Nextworks | Giertych M.,PSNC | And 3 more authors.
Computer Networks | Year: 2014

Network performance in terms of throughput, latency, packet loss or jitter significantly influences user's quality of experience of cloud applications. Network services impact on cloud applications performance and this impact is even more significant when the cloud infrastructure spreads over different administrative domains, such as in a federated cloud or hybrid-cloud scenarios. Given this strong coupling between cloud application performance and network performance there is great value to be gained by supporting advanced controlled networking functionalities between distributed cloud infrastructures. These functionalities would be useful to the Future Internet (FI) experimentation community as well as future production clouds. This paper describes an architecture and a set of procedures to interconnect a multi-cloud environment with advanced facilities for controlled networking. This integration allows the provisioning of customized network functions and services in support of experiments running in a multi-cloud test-bed. The possibility to control the network connectivity is a key feature to provide better performance for the experimenters' cloud applications. We focus on the details of federating three advanced networking facilities with the BonFIRE multi-cloud environment. These three networking facilities are: FEDERICA, which supports controlled routing; GÉANT's Bandwidth-on-Demand service and OFELIA that uses OpenFlow to provide Software Defined Network functionalities. The interconnections with FEDERICA and GÉANT are already active, while OFELIA is envisaged as future work for a third facility to interconnect. © 2013 Elsevier B.V. All rights reserved.

Hardt M.,Karlsruhe Institute of Technology | Jejkal T.,Karlsruhe Institute of Technology | Campos I.,CSIC | Fernandez E.,CSIC | And 5 more authors.
Computing and Informatics | Year: 2012

This paper describes the architecture for transparently using several different Cloud Resources from with the graphical Kepler Workflow environment. This architecture was proven to work by implementing and using it in practice within the FP7 EUFORIA project. The clouds supported axe the Open Source cloud Open-NEbula (ONE) environment and the commercial Amazon Elastic Compute Cloud (EC2). Subsequently, these clouds are compared regarding their cost-effectiveness, which covers a performance examination but also the comparison of the commercial against a scientific cloud provider.

Aravanis A.I.,Synelixis Solutions Ltd. | Voulkidis A.,Synelixis Solutions Ltd. | Salom J.,Catalonia Institute for Energy Research IREC | Townley J.,UTRCI | And 5 more authors.
2015 IEEE Globecom Workshops, GC Wkshps 2015 - Proceedings | Year: 2015

Recent advances in energy efficient data center (DC) design in combination with predictive optimization techniques have led to the emergence of DCs that are able to adapt their operation to their environment. Current approaches towards assessing the operational efficiency of the DCs are mostly static, disregarding the flexibility aspects of modern DCs, also failing to reflect the effects of DC optimization on their sustainability profile. In this context, eight EU research projects have joined forces to present new metrics allowing for the evaluation of DC flexibility as well as of the effects of DC optimization to their general operational efficiency. The present paper presents these metrics along with the verification methodologies employed for the evaluation of the obtained measurements and results. © 2015 IEEE.

Grammatikou M.,National Technical University of Athens | Marinos C.,National Technical University of Athens | Demchenko Y.,UvA Amsterdam | Lopez D.R.,RedIRIS | And 2 more authors.
Proceedings - 2011 3rd IEEE International Conference on Cloud Computing Technology and Science, CloudCom 2011 | Year: 2011

Cloud computing has become a common technology for provisioning infrastructure services on-demand. Modern Cloud platforms can provide cloud-based applications, software, deployment platforms, or general infrastructure services that may include both computational and storage resources. However existing Cloud provisioning models are based on proprietary solutions and don't allow the combination of services from different providers and/or user legacy application that usually are present in user home organizations or campus networks. This paper introduces GEMBus (GEANT Multi-domain Bus), a service-oriented middleware platform that allows flexible services composition, and their on-demand provisioning and deployment to create new specialized task-oriented services and applications. GEMBus is built upon state-of-the-art Enterprise Service Bus (ESB) technologies and extend them with new functionalities that allow dynamic component services deployment, composition and management. The current paper discusses the general case for integration of Service-Oriented Architecture (SOA) principles and technologies with the provision and deployment mechanisms of Cloud-based platforms to support on-demand infrastructure services provisioning. It describes the Composable Services Architecture (CSA) that provides a general framework for GEMBus services design and operation. The paper also presents the current GEMBus implementation status and discusses how it can be applied as a general SOA platform for Cloud-based service provisioning. Finally, it discusses the practical use case of the federated network monitoring service that can be used as integration component in creating/building GEMBus based Cloud infrastructure services. © 2011 IEEE.

GASTONIA, N.C., Oct. 31, 2016 /PRNewswire/ -- PSNC Energy (PSNC), a wholly owned subsidiary of SCANA Corporation (NYSE: SCG), announced today, that in an unanimous decision, the North Carolina Utilities Commission has issued an Order in PSNC's general rate case application authorizing...

Santuari M.,CREATE-NET | Doriguzzi-Corin R.,CREATE-NET | Gerola M.,CREATE-NET | Salvadori E.,CREATE-NET | And 8 more authors.
Proceedings - 2014 3rd European Workshop on Software-Defined Networks, EWSDN 2014 | Year: 2014

Nowadays, several devices supporting OpenFlow versions beyond 1.0 are available on the market. However, existent OpenFlow testbeds like OFELIA do not provide support for such devices, therefore it is hard for the research community to explore and test the newest functionalities of the protocol in realistic environments. This paper gives a short overview of two different architectures investigated within the FP7 ALIEN project that both aim to overcome this limitation within the OFELIA experimental facility. © 2014 IEEE.

Hazlinsky M.,CESnet | Pietrzak B.,PSNC | Farina F.,GARR | Sobieski J.,NORDUnet | Szegedi P.,TERENA
SDN and NFV: Next Generation of Computational Infrastructure - 2014 International Science and Technology Conference - Modern Networking Technologies, MoNeTec 2014, Proceedings | Year: 2014

The last decade has produced many important advances in virtualization of IT infrastructure. Virtual Machines, once a novelty, have become an integral component of production web applications and the foundation for cloud services. Virtual Circuits have been around even longer and are now seeing the deployment of multi-domain provisioning services. The conventional core internet routing and switching platforms are giving way to virtualized facilities including virtual routers, virtualized network functions, and even virtual switching and forwarding fabrics. © 2014 IEEE.

Sobieski J.,NORDUNet | Farina F.,GARR | Naegele-Jackson S.,Friedrich - Alexander - University, Erlangen - Nuremberg | Kramaric K.,NORDUNet | And 2 more authors.
Proceedings - European Workshop on Software Defined Networks, EWSDN | Year: 2015

This paper describes a new feature of the GEANT Test beds Service that allows the programmability of external interfaces with DSL based topology descriptions. This feature is now available in version 2.0, but will be even more extended in the next version of GTS with the automatic provisioning of multidomain federated resource pools. The demo will show the users on how to work with test beds created in GTS and how to be able to develop test beds that not only connect GTS resources, but also include external domains. © 2015 IEEE.

Loading PSNC collaborators
Loading PSNC collaborators