Satisfied customer CRP Gabriel Lippmann returns to ClusterVision to design, build and manage a new turn-key cluster extension project, delivering additional capability and features to their extended research and development community.
Belvaux, Luxembourg, 21 June 2012 — ClusterVision, Europe’s dedicated specialist for HPC cluster solutions, has announced the completion of an HPC cluster extension at the Centre de Recherche Public Gabriel Lippmann (CRP-GL), one of Luxembourg’s leading research and development facilities. The newly extended cluster provides high performance computing facilities for an extended community of research and commercial development partners, enabling applications in fields such as climatology, medical research, and manufacturing.
Located in Belvaux , in south-western Luxembourg, the Centre de Recherche Public Gabriel Lippmann (CRP-GL) is a public establishment devoted to the advancement of applied scientific research and technological development. Named in honor of Professor Gabriel Lippmann, Nobel Prize winner for Physics in 1908, CRP-GL has focus in 3 major areas :
- innovative materials technology, in particular nano-technologies and instrumental development,
- sustainable management of natural resources,
- technologies for the information society.
ClusterVision has been providing HPC cluster systems, software and services to CRP-GL since 2010 when, after a successful competitive tender process, they were awarded the contract to build and install the facility’s first supercompute cluster.
This initial cluster system comprised 32 compute nodes, containing 128 AMD Shanghai 8376HE processors, a total of 512 cores. The Supermicro quad-socket servers were housed and cooled in a Rittal racking system, with interconnect technology provided via Giga-Ethernet and a Mellanox Infiniband DDR network. The system also included 80TB of usable Supermicro storage, shared with NFS. The cluster’s on board provisioning and operation was managed by version 5.0 of Bright Cluster Manager, from Bright Computing.
Opened in April 2011, and reaching a peak performance of around 3.4 TFlop/s, the initial CRP-GL cluster system provided compute power to 18 different research and development projects, in areas such as environmental modelling and materials science.
In October 2011, pleased with the combined systems, software and services provided for their initial system, CRP-GL approached ClusterVision again, this time directly, in order to discuss their proposed cluster extension project. In addition to extending and further customising their compute capability, CRP-GL were keen to build upon the solid base of technology and management software, and the quality of user support services already in place.
Being involved at an early stage of the design process enabled the ClusterVision team to gain a thorough understanding of CRP-GL’s expanded requirements, and to recommend an extension design which would provide both a step-change in compute capacity, and one which could be quickly and harmoniously integrated with CRP-GL’s existing facility.
To meet these challenges, ClusterVision designed a cluster extension comprising 24 additional compute nodes and 2 redundant master nodes. 48 Intel X5650 processors, a total of 288 cores, were added to the existing system, this time delivered in Dell C6100 servers, again housed and water cooled by a Rittal enclosure system.
Several other areas of improvement were also built in to the extended system. The Mellanox DDR interconnect was upgraded to a QDR switched fabric InfiniBand, giving performance improvements in both signal rate and latency, and the storage was moved from standard NFS to Fraunhofer’s FhGFS parallel file system. Both items significantly increased the bandwidth needed for the I/O intensive post-processing and visualisation requirements of CRP-GL’s applications. To complete the enhancements, the system management software was upgraded to Bright Cluster Manager version 5.2, at the time the latest available version.
On-site installation of the cluster extension began in March 2012, with completion and final acceptance just 2 months later, with the project being delivered as a complete turn-key system, including training and on-going support.
Now fully operational, the CRP-GL cluster extension provides a powerful HPC capability both for the CRP-GL and its external customers. End-user research applications being run on the cluster cover a wide range of interests, including regional climate modelling, Monte-Carlo stochastic simulation of hydrological processes, advanced protein modelling, and molecular dynamics. They also include commercial applications in material science and automotive manufacturing.
“Having set up the initial HPC platform at our institute, we subsequently noted a steady increase in research projects and external customers using the platform and causing a demand for further compute and storage capacities. The successfully completed extension of our platform by ClusterVision to more than 6 TFlop/s compute and some 140 TB storage capacity helps us to satisfy these needs and to foster research carried out at CRP-GL by internal researchers and external customers,” Mr. Christian Anese, Head of IT and Organisation and Mr. Markus Jonas, System Administrator HPC platform CRP-GL, hpc-support [at] lippmann [dot] lu.
“Based on the success of our initial installation, we were delighted to be invited to return to CRP-GL to design and install their cluster extension. Working with the CRP-GL team from the early stages of their thought process gave us a real opportunity to create a highly tailored solution, building on their existing investment, while also including some innovative new hardware systems and software features,” Mr. Christopher Huggins, Director of Sales & Marketing, ClusterVision.
Centre de Recherche Public Gabriel Lippmann
About the Centre de Recherche Public Gabriel Lippmann
CRP Gabriel Lippmann focuses on three major axes: innovative materials technology, in particular nano-technologies and instrumental development, sustainable management of natural resources and technologies of the information society. Research work is carried out by some 200 researchers, teacher-researchers and 3rd year trainees, within three research departments:
- Cooperative IT based on NTIC, modeling of the information systems, architectures of the information systems software engineering and visualization of information are the fields of competence of ISC (IT, Systems and Collaboration Department)
- The study of ecosystems, hydro-climatological monitoring, application of agro-biotechnologies in the sectors agricultural and forest, durable management of the territory, nutrition and toxicology are the competence of EVA (Environmental and Agro-biotechnologies Department)
- Characterization of materials, surfaces and interface, characterization of the nanomatériaux, development of innovating surface treatments, design and development of new instrumentations and new equipment in the fields of the analysis of materials and the surface treatment are the competence of SAM (Science & Analysis of Materials Department)
Since its creation by grand ducal regulation in July 1987, the CRP Gabriel Lippmann has adopted an original approach, focusing on equilibrium between the criteria of economic utility and scientific quality rather than giving priority to one of them. As such, it can continue its activities involvement in a partnership with national companies and administrations, while also successfully carrying out several innovative research projects with European as well as national funding.
ClusterVision specialises in the design, deployment and operation of High Performance Compute (HPC) clusters. Compute clusters are used by pioneering organisations in academia and industry to run high-intensity computing applications in fields such as scientific research and development, manufacturing, healthcare and finance. By combining cutting-edge hardware and software components with a range of customised professional services, ClusterVision helps its customers create top-quality, efficient and reliable HPC solutions. In addition to systems technologies from leading manufacturers, ClusterVision's solutions typically include a range of HPC software components, such as easy to use cluster provisioning, management and monitoring. ClusterVision offers a full portfolio of professional services, covering the total cluster lifecycle - from system design, assembly and certification, to operational management, support, and training. With a background in applied scientific research, and practical experience in a wide range of HPC technologies, the ClusterVision team has designed and built some of the largest and most complex computational, storage and database clusters in Europe.
For more information
Dr. Tony Kent
Marketing & PR Manager
Tel: +44(0)7795 44 1655
pr [at] clustervision [dot] com