Bandwidth on Demand Services for European Research and Education Networks

Campanella, M.; Sevasti, Afrodite; Krzywania, R.; Stamos, Kostas; Reijs, Victor; Wilson, Dave; Tziouvaras, Chrysostomos

doi:10.1109/bod.2006.320801

Cited by 10 publications

(9 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many network resource reservation systems have been developed and deployed (Campanella et al 2006;Guok and Robertson 2006;Johnston, Guok, and Chaniotakis 2011;Riddle 2005;Zheng et al 2005;Sobieski, Lehman, and Jabbari 2004). However, existing systems either are inefficient, or cause private information to be exposed.…”

Section: Related Workmentioning

confidence: 99%

“…The key to supporting these applications is the ability to provide guaranteed network resources (i.e., bandwidth) for performance predictability (Mogul and Popa 2012). As such, many network resource reservation systems are developed and deployed (Campanella et al 2006;Guok and Robertson 2006;Johnston, Guok, and Chaniotakis 2011;Riddle 2005;Zheng et al 2005;Sobieski, Lehman, and Jabbari 2004). However, because of the underlying networks' concern of revealing sensitive information, existing reservation systems do not provide applications with an interface to access information of the underlying network infrastructure (e.g., topology, links' available bandwidth).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Optimizing in the Dark: Learning an Optimal Solution through a Simple Request Interface

Xiang

Aspnes

et al. 2019

AAAI

View full text Add to dashboard Cite

Network resource reservation systems are being developed and deployed, driven by the demand and substantial benefits of providing performance predictability for modern distributed applications. However, existing systems suffer limitations: They either are inefficient in finding the optimal resource reservation, or cause private information (e.g., from the network infrastructure) to be exposed (e.g., to the user). In this paper, we design BoxOpt, a novel system that leverages efficient oracle construction techniques in optimization and learning theory to automatically, and swiftly learn the optimal resource reservations without exchanging any private information between the network and the user. We implement a prototype of BoxOpt and demonstrate its efficiency and efficacy via extensive experiments using real network topology and trace. Results show that (1) BoxOpt has a 100% correctness ratio, and (2) for 95% of requests, BoxOpt learns the optimal resource reservation within 13 seconds.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Optimizing in the Dark: Learning an Optimal Solution through a Simple Request Interface

Xiang

Aspnes

et al. 2019

AAAI

View full text Add to dashboard Cite

show abstract

“…In Europe, there are National Research and Education Networks (NRENs) in each country which are interconnected by the GEANT network (the provider organization is called DANTE). Related to the evolution of the GEANT network, research and service activities have been undertaken in the GN2 project (2004 to 2009, [4]) and are continuing in the new GN3 project (2009 to 2013). In GN2/GN3, a software called AutoBAHN ("Automated Bandwidth Allocation across Heterogeneous Networks", [5]) has been developed as a BoD solution that is interoperable with other software such as the DRAGON [6] and OSCARS [7] systems from Internet2/ESnet.…”

Section: Bandwidth-on-demand Scenario For the European Academic Nmentioning

confidence: 99%

Investigating the economic feasibility of Bandwidth-on-Demand services for the European research networks

Hanemann

Hausheerx

Reiche

et al. 2010

2010 IEEE/IFIP Network Operations and Management Symposium Workshops

View full text Add to dashboard Cite

In the recent years, several technical solutions have been developed that allow Bandwidth-on-Demand services to be offered by service providers. In the context of the European academic networks, the technical feasibility of such services has been demonstrated on several occasions. The move from a research activity to a production service presents further requirements in addition to the overcoming of technical challenges. Most notably, apart from organization of day-today operations, financial issues have to be solved. These relate to sharing the costs of a service that is delivered jointly by independent organizations. The main cost factors are operations staff as well as network resources that are purchased for ondemand use. Moreover, the financial issues also include pricing models for use of the service use as well as the distribution of the revenues among the participating organizations. The present paper considers this fundamental BoD scenario together with key challenges in detail, before discussing the general approach for specifying a cost/pricing model.

show abstract

“…To achieve this, multiple VLAN's on research networks such as those employing AutoBAHN [6] and Internet2's ION [7] are stitched together to achieve the layer 2-connectivity. In this test, for each combination of testbeds that support stitching the following steps are executed after creating a new slice on the central SA:…”

Section: Request the Supported Protocols Of The Testbed'smentioning

confidence: 99%

Health monitoring of federated Future Internet experimentation facilities

Walcarius

Vandenberghe

Vermeulen

et al. 2014

2014 European Conference on Networks and Communications (EuCNC)

View full text Add to dashboard Cite

Abstract-The federation of Future Internet testbeds as envisaged by the Fed4FIRE project is a complex undertaking. It combines a large number of existing, independent testbeds in a single federation, and presents them to the experimenter as if it were a single infrastructure. Operating and using such an infrastructure requires a profound knowledge of the status of the health of the underlying independent systems. Inspired by network monitoring techniques used to operate the Internet today, this paper considers how a centralized health monitoring system can be set up in a federated environment of Future Internet Experimentation Facilities. We show why it is a vital tool for experimenters and First Level Support in the federation, which health monitoring information must be captured, and how this information can be displayed most appropriately.

show abstract

Bandwidth on Demand Services for European Research and Education Networks

Cited by 10 publications

References 7 publications

Optimizing in the Dark: Learning an Optimal Solution through a Simple Request Interface

Optimizing in the Dark: Learning an Optimal Solution through a Simple Request Interface

Investigating the economic feasibility of Bandwidth-on-Demand services for the European research networks

Health monitoring of federated Future Internet experimentation facilities

Contact Info

Product

Resources

About