Infrastructure‎ > ‎

Testbeds


BonFIRE brings together leading testbed providers to offers a pan-European federation of test beds for large scale experimentation of cloud and service technologies. Each testbed provider offers resources using an Infrastructure-as-a-Service delivery model which are then provisioned for experimentation using BonFIRE methodologies and tools.

Most of the testbed providers offer both permanent and on-request resources. Permanent resources target experimenters who are tackling smaller scale problems and require fast access. On-request resources are for larger-scale experiments and will be made available on request for specific experiments. Moreover, the on-request functionality also allows you to reserve a physical machine (or cluster of physical machines) exclusively to your experiment.

The set of testbeds currently offered by BonFIRE is shown in Figure 1, which comprises EPCC, HP, iMinds, Inria, USTUTT, PSNC and Wellness Telecom (WT). Detailed information about the infrastructure offerings is maintained [here].

BonFIRE Infrastructure Overview

Figure 1: BonFIRE Infrastructure Overview

Deploying and accessing resources on the different testbeds is done in a seamless manner via the BonFIRE API, which is based on OCCI. Since the availability of public IP addresses varies across the testbeds, BonFIRE implements a WAN to which all VMs can be connected. SSH Gateways are in place at every testbed, allowing experimenter to connect to the VMs that are not publicly accessible, using SSH. Moreover, each testbed is connected via the BonFIRE WAN, which means you can access VMs on any testbed via a single SSH Gateway.

EPCC

EPCC is a leading European centre of expertise in advanced research, technology transfer and the provision of supercomputer services to academia and enterprise. EPCC provides world-class computing facilities for scientists and collaborates with researchers in areas ranging from particle physics to cosmology, novel materials to combustion and climate modelling to energy. Our work not only benefits scientific research, but generates new techniques and approaches that are taken up by the industry.

We have recently replaced our production infrastructure, greatly increasing capacity. Our hardware is based on standard Dell servers, using the latest multi-core processors and RAID disk arrays to provide a robust platform for VM deployment. Our production infrastructure consists of:

1 server Intel Xeon E5620 (2x4 cores) 2x 148GB + 12x 2TB
2 workers AMD Opteron 6176 (4x12 cores, 128GB RAM each) 2x 600GB disk Gigabit Ethernet


Inria

INRIAEstablished in 1967, Inria is the only public research body fully dedicated to computational sciences in France.

Combining computer sciences with mathematics, Inria’s 3,400 researchers strive to invent the digital technologies of the future. Educated at leading international universities, they creatively integrate basic research with applied research and dedicate themselves to solving real problems, collaborating with the main players in public and private research in France and abroad and transferring the fruits of their work to innovative companies.

The researchers at Inria published over 4,800 articles in 2010. They are behind over 270 active patents and 105 start-ups. In 2010, Inria’s budget came to 252.5 million euros, 26% of which represented its own resources.

Inria operates Grid'5000, a scientific instrument designed to support experiment-driven research in all areas of computer science related to parallel, large-scale or distributed computing and networking and present in 9 sites in France and 1 site in Luxembourg. It also operates one BonFIRE's testbed.

Inria's testbed on BonFIRE has 8 worker nodes permanently available to BonFIRE users and can expand over the 160 nodes of Grid'5000 located in Rennes. When using on-request resources of Grid'5000, BonFIRE users have a dedicated pool of machines that can be reserved in advance for better control of experiment conditions, but nevertheless accessible using the standard BonFIRE API.


USTUTT

Founded in 1829, the foremost aim of the Universität Stuttgart (USTUTT) in research and teaching lies in the fields of engineering and the natural sciences.The university with its 140 institutes distributed over 10 faculties, 5,000 employees, and about 20,000 students - among them 3,000 young men and women from all over the world - has a strong impact in Baden-Württemberg, Germany’s traditional high-tech region.

Rechenzentrum Universität Stuttgart (RUS) is the university's computing center and acts as a central service facility for computing, networking and information services and has initiated and manages the state science network ‘BelWue’. High Performance Computing Center Stuttgart (HLRS), is operating the first German National supercomputer center for university research and is visible in this role at an international scale. Of the centre’s more than hundred persons staff about fifty are engaged in State, Federal, European and international projects. Both RUS and HLRS have great experience in national and international research projects such as MAMS, Akogrimo, Daidalos and IRMOS, to name only a few.

USTUTT provides permanent resources for the BonFIRE testbed, which currently amount to 14 worker nodes. An extension of this permanent infrastructure is currently underway. Furthermore, USTUTT offers the possibility of using supercomputer resources.

iMinds

iMindsThe iMinds organisation, formally known as IBBT (Interdisciplinary Institute for Broadband Technology) is an independent research institute founded by the Flemish government to stimulate ICT innovation, and hosts the iLab.t technology and knowledge centre to support experimental multi-disciplinary research for Future Internet technologies.

The Virtual Wall in iLab.t is a generic test and emulation environment for advanced network, distributed software and service evaluation, supporting scalability research. It consists of 100 nodes, interconnected via a non-blocking central switch, and a display wall of 20 monitors for experiment visualization.

The iMinds Virtual Wall supports experimenters with a fully controlled network environment, where various network topologies can be defined and link capacities and impairments (packet loss rate, delay, background traffic) can be adapted on-the-fly.


PSNC

PSNCThe Poznan Supercomputing and Networking Centre (PSNC) is an active research and development centre involved in numerous national and European projects in the field of networking, media, clouds and services. It is also responsible for the development and management of the national research & educational network (PIONIER) in Poland.

PSNC provides a testbed of up to 3 nodes where each platform consists of dual Intel Xeon E5645 (6 cores) CPUs with 1,2TB storage and 4x 1GbE connectivity. The internal network supports up to 10GbE links with the 1 GbE uplinks to the Internet.

The PSNC site not only increases the BonFIRE computational power but it will also support experimenters with a controlled network environment to provide opportunities for conducting experiments with strong requirements on the underlying network infrastructure. This scenario will be implemented in cooperation with external project GÉANT and its Bandwidth on Demand service (http://bod.geant.net). Through this feature users may request dedicated connection between two sites where experiment resources are set up. This dedicated link will provide guaranteed bandwidth and stable isolated connection across public networks. This is an attractive feature that can be utilised by network aware experiments, e.g., where huge amounts of data have to be exchanged in predictable period of time between compute nodes in different locations. Initially this network scenario will be available between EPCC and PSNC testbeds.


Wellness Telecom

Wellness Telecom
Wellness Telecom is a young Information and Telecommunication Technologies SME located in Seville (Spain). It endeavors to offer innovative solutions using cutting edge technologies at national and international levels. Wellness Telecom's mains specialization areas are based on horizontal IT technologies applied to different action fields: Health, Smart City, Cloud, Energy, Environment, etc.

Wellness strengthens its cloud business line with the creation of Squada, the new company division focused on Cloud Computing and IT systems. Through this brand, it offers ISP (Internet Service Provider) and CSP (Cloud Service Provider) services. Wellness Telecom uses VMware solutions for cloud computing. Overall, VMware stands out as the cloud leader – and in particular, a leader in enterprise hybrid cloud solutions – due to the reach and maturity of its on-premises and off-premises offerings: IaaS solutions, its broad service provider ecosystem for IaaS, the industry-leading vFabric PaaS suite, Cloud Foundry for open standardsbased PaaS, a broad portfolio of virtualization and cloud management solutions, and aggressive development of cross-cloud enabling tools and interfaces to bring them all together. In a recent Gartner market analysis of cloud services providers, four out five top service providers work with VMware technologies.

Wellness Telecom provides 8 dedicated nodes as permanent resources with a total pool of 120 GHz (Intel® Xeon® Processor E5-2640 2.50 GHz 95W 6C/15MB Cache/DDR3 1333MHz). Each node has 8GB of RAM memory (DDR3-1600-MHz RDIMM/PC3-12800/dual rank/1.35v) with a total of 3TB of storage capacity in the Storage Area Network (SAN).