Partner-Report – Meet SESAME Net

The Partner Report is a multi-part reportage to introduce our network members. Each newsletter will showcase some of our partners, and highlight some of their most recent activities. This edition will feature the Supercomputing Centre of Galicia.

The Supercomputing Centre of Galicia (CESGA Foundation) is the centre for computing, high performance communications systems, and advanced services of the Galician Scientific Community, the University academic system, and the National Scientific Research Council (CSIC).

CESGA Foundation is a non-profit institution whose trustees pertain to the Regional Government of Galicia and to CSIC. CESGA has been one of the main HPC infrastructures in Spain for the last quarter century.

Foundation CESGA has the mission to contribute to the advancement of Science and Technical Knowledge by means of research and application of high performance computing and communications, as well as other information technologies resources, in collaboration with other institutions, for the profit of society.

In general, the purposes of the Foundation are all those that promote the research in and use of high performance computing, advanced communications and development of information and communications technologies as an instrument for sustainable socioeconomic development, devoting special attention to the cooperation between research centres, whether public or private, and the productive sector.

On top of acting as an advanced services centre for RTD institutions, CESGA has participated in numerous European Research Framework Programme Projects. For the past 15 years CESGA has carried out extensive work to help SMEs benefit from the exploitation of advanced computing technologies.

Thus, CESGA has displayed extensive activity both with research groups and enterprises in the following areas:
– Promoting  and participating  in research and technological development projects.
– Providing computing capacity, communications and technical support to its users by means of the equipment available at the Foundation.
– Collaborating  in the transfer of research results in the area of computing between public research centres and private companies.
– Boosting and promoting  cooperation between institutions at the national and international level.
– Contribute to the development and strengthening of the competitive capacity of the Galician and CSIC research communities.
– Collaborating with institutions, companies and people in any matter related to the object of the Foundation, in the terms established in the corresponding contracts or cooperation agreements.
– Promoting and collaborating in the organisation of courses, seminars and meetings related to innovation by means of computing and communications.
– Promoting and disseminating those activities and results considered of interest for the Foundation.
– Collaborating with other supercomputing centres for an effective use of the technologies.
– Searching for and obtaining resources for the development of its activities.
– Tend to the technological needs of the entities and companies that so require it in the area of the supercomputing.
CESGA is part of the Spanish Supercomputing Network and as such it is recognized by the Spanish Ministry for the Economy and Competitiveness as a Singular Scientific Technological Installation (ICTS).

Successful collaboration with Flemish Supercomputer Center enables Bekaert to model wire manufacturing

Bekaert is a world market and technology leader in steel wire transformation and coating technologies. Headquartered in Flanders, it is a global company with almost 30.000 employees worldwide. Bekaert is a prime supplier of steel cord products for tire reinforcement and steel fiber for concrete reinforcements. But it offers a very wide range of advanced steel wire products for other applications to many sectors.

‘Calculations are more economic than physical tests’

For several of its steel wire transformation processes, Bekaert relies on supercomputing. Peter De Jaeger, Team leader Modelling at Bekaert, explains:
“We use computational models to predict the behavior of steel wire during different transformation processes, e.g. wire rolling, wire drawing, rope manufacturing, … Such calculations are much faster and more economic than relying on physical tests on proofing samples. In this way, supercomputing allows us to drastically enhance the speed of product development and improve the development processes.”

‘Supercomputing drastically enhances product development’

“But this modeling approach requires proper software implementation and ample computational resources.” For both, Bekaert relies on the services of the Flemish Supercomputer Center.

The Flemish Supercomputer Center (Vlaams Supercomputer Centrum – VSC, located in Belgium) is a partnership between the five Flemish universities and their university associations, and is managed by the national funding agency, Research Fund – Flanders (FWO). This consortium brings together knowhow in scientific and technical computing (including high performance computing, high throughput computing, cloud computing and data processing) in Flanders. It houses infrastructure in four hubs: the data centers of the universities of Antwerp, Brussels, Ghent and Leuven.

‘The VSC team offers world-class supported computational infrastructure’

The VSC encourages the use of scientific and technical computing in the Flemish academic and industrial landscape. To this end, it offers infrastructure, training and services. In addition, VSC acts as a lever to promote the importance of scientific and technical computing and its added value to society.

The collaboration with Bekaert developed as part of VSC’s mission to actively promote scientific and technical computing in Flemish industry and foster the exchange of ideas and expertise between research institutions and industry.

For Bekaert, the VSC support team specifically installed a tailor-made version of the OpenFOAM-extend software and made this software available on all clusters within the VSC network, including the first Flemish Tier1 supercomputer, the fastest Belgian compute cluster at that moment. All Bekaert’s computational models now run in an efficient manner on VSC’s infrastructure.

Peter De Jaeger: “The VSC team offers world-class supported computational infrastructure which makes high performance computing feasible from an industry perspective.”

Funding opportunities for SMEs: Fortissimo 2 project Calls for Proposals

FORTISSIMO2

The Fortissimo 2 project is funded under the European Commission’s Horizon 2020 Framework Programme for Research and Innovation and part of the “ICT Innovation for Manufacturing SMEs (I4MS)” action. Building on the Fortissimo project (funded under the Seventh Framework Programme) established in 2013, Fortissimo 2 contributes to the following technology areas:

  • HPC Cloud-based modelling, simulation and analytics services for modelling multiple interconnected phenomena.
  • integrating multiple tools across the process chain.
  • exploiting the dynamic availability of Big Data.
  • integrating novel mobile interfaces for data management and decision support.
  • achieving real-time response (application experiments), and addressing comprehensively security and privacy issues at all levels.

The core objective of Fortissimo 2 is to promote the uptake of advanced modelling, simulation and data analytics services by European small and medium enterprises (SMEs) in the manufacturing sector. Adopting latest ICT technology is essential to providing innovative solutions to manufacturing challenges, leading to new and improved products and services.

To achieve its objective, a set of application experiments will be carried out in the course of the project. Fortissimo 2 experiments must address engineering and manufacturing needs in the areas of high-performance data analytics or the modelling and simulation of coupled physical processes. The first batch of experiments has already begun in early 2016; two additional sets of experiments will be selected through open Calls for Proposals (CfP).

Fortissimo 2 will organise two CfPs for additional application experiments. “Call-1” will accommodate experiments to be executed for an 18-month period commencing in November 2016. “Call-2” is expected to be launched in September 2016, leading to another set of 18-month-long experiments starting in May 2017.

The expectations for the proposed experiments are that they should:

  • be complementary to those already included in Fortissimo 2 and to those from the original Fortissimo project,
  • contain all those actors in the value chain necessary for the realisation of services meeting the end-users’ engineering and manufacturing needs, and
  • use the (distributed) HPC resources already offered within the Fortissimo infrastructure to solve real world problems involving high performance data analytics or coupled simulation of physical phenomena.

The new application experiments should provide business relevant investigations and demonstrations of coupled simulation services or Big Data analytics in the Fortissimo HPC Cloud that have the potential to create business benefits for manufacturing SMEs. Priority will be given to experiments that are driven by the requirements of first-time HPC users and to those which involve SMEs and mid-caps. The business-relevance of the application experiments is essential.

Proposed experiments should include all participants necessary for the experiment that may include HPC and Big Data experts, HPC Centres or ISVs either already included within or external to the Fortissimo 2 consortium. Experiments will employ the Fortissimo HPC infrastructure using the HPC Centres already involved in the project.

Detailed instructions for proposal submission, together with information about the evaluation criteria to be applied, will be provided online at: http://www.fortissimo-project.eu/

HPC-SME-Success Story

healthincode_image005_logo

Our 3rd newsletter presents the successful collaboration between a HPC centre (CESGA) and a Spanish SME that faced a typical challenge of a sudden and abrupt increase of their computation and storage needs.

Many SMEs simply cannot invest in significant hardware resources to cope with sudden bursts in demand, nor have the relevant in-house expertise, to address this type of challenge. But one solution that many SMEs may not be aware of is that there are many European HPC centres with SME engagement programmes that can help SMEs to “fill in the gap” in a very short time with little financial investment.

healthincode_image001Our case study is about a Spanish SME, “Health in Code”, that is developing a genetic diagnostics technology to better treat patients with heart diseases. This SME was established in 2006 as a Spin-off of the University of “A Coruña” (Galicia – Spain), focused in genetic diagnosis of inherited cardiovascular diseases (a leading cause of sudden death, especially among young individuals). It has a multidisciplinary team and a global customer base: Spain, USA, UK, Ireland, Denmark, Italy, Israel, Greece, Germany, Russia, Australia, Argentine, Brazil, Denmark, Qatar, etc. While some cardiovascular diseases have genetic causes, clinical symptoms are heterogeneous that makes it very difficult to predict their evolution, so a genetic diagnosis can be useful in dispelling doubts about the presence or absence of genetic mutations (and its clinical relevance), which then informs tailored treatment that mitigate risks. This is where “Health in code” performs their specialized work and provides their expertise.

In order to extract more informative knowledge from their customers’ genetic profiles “Health in Code” have upgraded their DNA sequencer to take advantage of the latest high-throughput sequencing technologies. Their new Illumina HiSeq sequencer could generate 65,000 times more information than the previous one so they had to look for a solution to manage a much larger amount of data. Each patient now generates 4Gb of data that must be analysed and compared to heterogeneous databases of genetic variations, diseases, patients, genes, families and medical articles, i.e. a knowledge base developed by “Health in Code”. This information processing involves an initial step to process raw data to obtain a list of variants, a second step to “clean” obtained information from the data relevant to the study in question, and then a third step to compare relevant information to the heterogeneous knowledge base, and finally the automatic generation of a complete clinical report for physicians. All these information processing and analyses require a diverse range of bioinformatics software and high-end computational resources in order to achieve results within reasonable times.

Computation and storage services provided by CESGA have successfully addressed and resolved the company’s challenges, demonstrating the centre as a reliable partner to achieve results with the desired accuracy, efficiency and speed. This kind of collaboration should not only be viewed as a short term engagement purely in the context of hardware resources “for rent”, it also demanded mutual knowledge transfer to keep up with “state of the art” technologies: the SME benefits from better understanding of the latest hardware and software technologies, whereas the HPC centre gains expertise in different domains where HPC is a powerful tool.

As a result of the collaboration with CESGA, “Health in Code” is evaluating the adoption of additional Big Data technologies (powered by CESGA resources) to continue this partnership. The strive towards more accurate results and improved diagnostics demonstrates that SME-HPC centre engagement is invaluable for both short and long term benefits.

Author CESGA

Partner-Report – Meet SESAME Net

cesgaThe Greek Research and Technology Network (GRNET S.A., http://www.grnet.gr) is a state–owned non-profit company, operating under the auspices of the Greek Ministry of Education – General Secretariat for Research and Technology. Its mission is to provide high–quality electronic Infrastructure services to the academic, research and educational community of Greece; and link these with global e-Infrastructures.

Among other activities, GRNET is strongly involved in pan–European High-Performance Computing Initiatives, being a core member of PRACE project series with a mandate from the Greek Ministry of Education. In this framework the HPC Centre has coordinated the HP–SEE project (“High–Performance Computing Infrastructure for South East Europe’s Research Communities”), which established the HPC infrastructure in South East Europe.
Recently (June 2015), GRNET, as the national supercomputing facility of Greece, has deployed the Greek national HPC Tier–1 centre ARIS (Advanced Research Information System). ARIS is built on IBM NeXtScale platform and Intel® Xeon® E5 v2 processors (Ivy Bridge). System theoretical peak performance exceeds 190 TFlops (trillion floating point operations per second). The system houses 426 nodes, offering a total of over 8500 processing CPU cores, interconnected using Infiniband FDR fabric, a technology offering very low latency and high bandwidth. The infrastructure is complemented with a 1 Petabyte (quadrillion bytes) high performance storage system of, which is built on the IBM Elastic Cloud – the successor of IBM’s Parallel File System (General Parallel File System – GPFS). Finally, the system includes software for developing applications such as compilers, scientific libraries and popular suites of scientific applications. Regarding security, the system offers full redundancy with each server to have dual power supply, and there is an integrated UPS installation in a separate room which ensures the uninterrupted operation of the server for 15 minutes, which is enough time for closing the system safely in the event of power failure. A highly qualified, dedicated team of computer engineers/scientists supports the cutting-edge technology system.

For cloud computing (photo below), GRNET operates Infrastructure as a Service ~okeanos via large datacenters (22 racks, 400+ servers, more than 9000 Virtual Machines active, 4 Petabytes of storage), and is/was involved in a number of core pan-European cloud projects such as StratusLab, CELAR, etc. GRNET is also developing Synnefo, the cloud software for ~okeanos.

GRNET has been involved in a series of training activities within all the areas of its expertise described above, and runs national and international training events for networks, HPC and distributed computing related topics. GRNET has been a member of the PRACE FP7 programme, and has run South East Europe seasonal schools, and coordinated the seasonal schools activity in the PRACE-2IP project.

Author GRNET

Queuing or not queuing – is interactive HPC an attractive service for SMEs?

queuingMany companies have been moving from batch HPC processing to nearly real-time data processing, with in-situ visualisation getting all the attention from SMEs as a new user-facing HPC service. In a nutshell, the idea behind the real-time visualization is to perform rendering operations during the execution process, so users could react or even steer computing HPC simulations at any stage, for instance check whether application parameters were set correctly or they have to be modified due to instability. Continue reading

HPC-SME-Success Story

virtual prototyping

Use Cases and success stories are a powerful tool to illustrate how companies from different areas of business can benefit from HPC technologies. In this newsletter, we present and describe the cooperation between IT4Innovation’s and BORCAD, a producer of railway and medical technologies.

Continue reading