E - I N F R A S T R U C T U R E R E F L E C T I O N G R O U P

Fall 2015 1

The importance of being earnest about the long tail of researchPage 10

e-IRG – NewsletterPaving the way towards a general purpose European e-Infrastructure

Highlights from the delegates meetingSince the publishing of the previous edition of this Newsletter, there was one delegates meeting, in Riga.Page 4

Data is lightobservations from the 6th RDA PlenaryPage 20

European Commission's Work Programme 2016-2017Focus on e-infrastructures and HPC from SME to exascalePage 6

Open Science perspective highlighted at e-IRG workshop in RigaPage 5

Conference reportPLAN-E MunichPage 17

Newsletter Fall 2015

e-IRG Workshop in RigaOpen Science & e-Infrastructures

Main conclusions from the e-IRG workshop on June 3rd 2015 in Riga, Latvia:

The funding of Open Access to scientific publications, research data, software, methods, educational material and other resources needs to be addressed, in particular the provision of support for access and storage/maintenance

New ways of measuring the impact of Open Science are required.

The creation of the European Open Science Cloud for Research needs to be addressed (as an incarnation of the e-Infrastructure Commons).

A framework to be able to share infrastructure costs is essential.

There needs to be a close working relationship with ESFRI to address the coordination process of Member States investment strategies in e-Infrastructures.

A joint ESFRI-e-IRG Working Group needs to be established.

The e-Infrastructure Commons has another dimension that is Global, EU, regional, institutional, etc. !

Read further at Page 13 >>

The Building Blocks of the e-Infrastructure CommonsThe Luxembourg edition of the e-IRG Workshop will be organised November 24-25, 2015 at the Campus Belval of the University of Luxembourg in Esch-sur-Alzette.Read further at Page 2 >>

European Commission has launched Digital4Science PlatformThe European Commission has launched the Digital4Science platform set-up to stimulate conversations on Excellence in Science and its activities around Open Science, e-infrastructures, Future & Emerging Technologies and the FET flagships. Read further at Page 11 >>

E - I R G N E W S L E T T E R

Fall 2015 2

e-IRG Workshop “The Building Blocks of the e-Infrastructure Commons” theme of Luxembourg e-IRG WorkshopThe Luxembourg edition of the e-IRG Workshop will be organised November 24-25, 2015 at the Campus Belval of the University of Luxembourg, in the main building, called the "Maison du Savoir" - House of Knowledge, located in Esch-sur-Alzette/Belval. The e-IRG Workshop will address the building blocks of the e-Infrastructure Commons.

Antoine Barthel will pronounce the welcome address as the local host of the workshop, followed by a political address from a representative of the Ministry of Luxembourg. Rudi Balling from LCSB will provide the keynote, followed by e-IRG Chair Sverker Holmgren who will introduce the theme of the workshop.

Session 1: Progress on e-Infrastructure CommonsThe first session with two speakers will be devoted to the evaluation of the progress of the e-Infrastructure Commons. Arjen van Rijn will present the view of e-IRG while Augusto Burgueño from the European Commission's DG Connect will talk about the Open Science Cloud.

After lunch, there will be break-out sessions to discuss the progress of the e-Infrastructure

Commons. Four to Six mixed groups of experts and users of networking, computing and data will start from the e-Infrastructure Commons Summary document to feed the discussion.

Session 2: National perspectives The second session with three speakers will highlight the national perspectives with regard to a "one-stop shop" for e-Infrastructure users. Jan Gruntorad and Jan Oppolzer, both from CESNET, will expand on the CESNET approach to e-Infrastructure users. David Salmon form JISC will talk about the e-Infrastructure convergence in the UK.

Session 3: Re-using existing infrastructureIn a third session, two speakers will specify the topic of re-using existing e-Infrastructures. Steven Newhouse from ELIXIR will present the existing projects in the ESFRI Roadmap. Kay Graf from KM3NET will highlight the projects that submitted proposals to the ESFRI Roadmap 2016.

The closing session of the first day will report on the results from the break-out session teamwork earlier on, followed by a discussion.

Session 4: Networking aspects The second workshop day will start with networking aspects. This session has three speakers. David Ferguson from the UK will present the user view on information assurance and Authentication, Authorisation and Identification aspects (AAI) in the health sector. Ann Harding from SWITCH will talk about the need for higher information assurance and AAI to support data-intensive research from the NREN perspective. John Dyer from GEANT will expand on AUP mapping and harmonisation and the implications of it for end-to-end federated service provision and Open Science Cloud.

Session 4: Market placeFotis Karayannis from e-IRGSP4 will start the second session with a view on a one-stop shop marketplace. This talk will be followed by a panel discussion, featuring Steve Cotter from GEANT, Steven Newhouse from Helix Nebula, Tiziani Ferrari from EGI.eu, Florian Berberich from PRACE, Per Öster from EUDAT, and Paolo Manghi from OpenAIRE.

The workshop will be closed by e-IRG Chair Sverker Holmgren, Antoine Barthel and Arjen van Rijn.

The detailed programme and information on registering at the e-IRG website. !

http://e-irg.eu/e-irg-workshop-nov-2015

Progress Report 2015Group of Senior Officials on Global Research InfrastructuresThis report sets out the progress of the activities conducted by the Group of Senior Officials (GSO) on global research infrastructures (GRIs) since the approval of the

GSO Framework by the G8 Science Ministers in July 2013.

Since the endorsement of such framework, the GSO has been proactively working in parallel along two lines, the first aimed at identifying concrete opportunities for collaboration based on a number of research infrastructures (RIs) proposed to the Group by the individual GSO members and, the second, aimed at identifying possible common grounds for the develop- ment of policies in the domain of Access, Data Management, Research Infrastructure Life Cycle and Evaluation, which the GSO was mandated to work on. !

http://knowledgebase.e-irg.eu/documents/243153/296831/Group+of+Senior+Officials+on+Global+Research+Infrastructures+Progress+Report+2015.pdf

Group of Senior Officials on Global Research InfrastructuresProgress Report 2015Meeting of the G7 Science Ministers8-9 October 2015

E - I R G N E W S L E T T E R

Fall 2015 3

Moving up to an Open Science CommonsAt the e-IRG Workshop, held in Riga, Latvia, on June 3, 2015, we had the opportunity to interview Sergio Andreozzi, Strategy and Policy Manager in the European Grid Infrastructure (EGI), on the topic of the Open Science Commons. The concept of Open Science Commons is very useful to the discussion of the e-IRG Workshop, according to Sergio Andreozzi. Before defining Open Science Commons, Sergio Andreozzi would like to define the two terms of which the concept is composed. Open Science refers to the process of opening the creation and dissemination of scholar knowledge to as many participants as possible. It is a concept that already existed in the 17th century with the birth of the journal systems and which goal it was to offer researchers a channel to disseminate the results of their work. It was a step towards opening science in terms of opening its results.

Nowadays, we can do a lot more thanks to the development of new technologies such as the Internet. Science has become digital. It is possible to think about large research collaborations, thousands of researchers who collaborate together, not only to disseminate research results but also to make science. The openness also involves the creation of scientific results. This ranges from generating ideas, defining possible research goals, to the gathering of data, access to open data that is possibly available on the Internet, to using large-scale infrastructures, tools and software to analyze data.

Open Science nowadays also involves a broad amount and types of resources. Sergio Andreozzi thinks about the need to share data. This involves resources such as scientific instruments, digital data, e-Infrastructures, access to large-scale computing capabilities - this could be commodity computing or supercomputing - and also services to manage data. Finally, it is very important to access expert people, the intellectual capital of experts who can support researchers so they can better execute their jobs.

The question is: How do we manage all these types of shared resources, in a way that their value for the researchers is maximalized? Here, the concept of the Commons enters

into play. The Commons is a way, an approach to govern shared resources within a community in a way that researchers feel engaged in defining the rules to govern the resources, to share them in a way that their value is maximized.

We can see the Commons already applied in a number of sources. For instance, in software we find Linux as a type of Commons where basically there is a number of defined rules from the community where software engineers can provide a contribution to a common database. There are rules to define how this is accepted, how to get results, etc.

Another type of Commons in the digital era is Wikipedia where a large amount of people contribute to build the common knowledge base. They feel engaged and there are clear rules for them to engage in order to develop this common shared body of knowledge.

In the infrastructure area, we see that the concept of Commons is applied to infrastructure resources such as GEANT. A few years ago, a high level expert group defined a roadmap to evolve the governance

bodies towards becoming a Communication Commons for the knowledge communities in Europe. Later the Commons concept was expanded towards other infrastructures.

e-IRG came up with the vision of an e-Infrastructure Commons where basically all the infrastructures are interoperable, integrated and easy to access for the research communities. With the Open Science Commons, we are taking a step forward, according to Sergio Andreozzi. We recognize that Open Science requires not only communication infrastructures but also other types of resources that need to shared and governed by the community. For instance, we have data, we have experts, people who can support researchers in doing their work - if you think, for example, about the concept of e-Science centres. All the types of interrelated resources that are needed for science can be managed as Commons.

We should not see the Open Science Commons as a unique gathering of all the resources but there are different types of Commons, each with their own governance, according to the principal of committed

participation of the users who can influence the rules to access the resources and so on. Once this is implemented, the vision should be that researchers from all disciplines have access to all the services they need and additional services such as knowledge expertise and instruments that enable them to collaborate and to achieve excellence in science but also that they feel engaged to govern these resources in defining the rules for further evolution.

To pick out an example in the area of data, we can see the Genome Data Commons which is the public availability of all data, coming from the Human Genome project, that was made available according to well-defined rules, stating when the data should be available and to what kind of groups.

In network computing, Sergio Andreozzi foresees the creation of a European backbone of tools and resources that come from the individual member states that contribute to the creation of a generic infrastructure on top of which different research infrastructures of the long tail can build their own research

platforms. This can be structured as a Commons if individual institutions of each country can contribute to a certain pool of resources or common infrastructures; and if they can offer these capabilities which answer to open standards and do not create artificial lock-ins; and if there is a clear governance defined at different levels, from the local to the national and to the European level, that enables the communities who use the resources - which could be ESFRI projects or the long tail representatives - to define and

participate in the evolution of this generic Commons Infrastructure in a way that this can follow their needs.

The standards can be seen as instrumental to create the Commons, to make the sharing of resources available, to enable, for instance, even the coexistence of resources coming from the private sector or from the public sector. As long as they are based on open standards, communities can have the freedom to participate and choose what best fits their needs without creating artificial enclosures or lock-in to a certain provider. The users can also participate in the trends and evolution of the services provided by the private or the public sector and can have a role in defining how these resources are being shared within the community.

With the Open Science Commons, Sergio Andreozzi believes that we are at the beginning of an interesting journey of applying the Commons principle to a wider set of shared resources, not just infrastructures but also other types of resources. !

http://go.egi.eu/oscd

E - I R G N E W S L E T T E R

4 Fall 2015

Highlights from the delegates meetingFour times a year, the e-IRG delegates, representing their countries gather for a working meeting. These meetings are organised in the country that at that moment holds the Presidency of the European Union. Since the publishing of the previous edition of this Newsletter, there was one meeting in Riga from which the minutes are already available. Although the meetings are only open for delegates to attend, there are public minutes published of each meeting.

These are available on the e-IRG web site. Because these delegates meetings are typical working meetings, a lot consists of reporting and discussing ongoing work. Hence when one meeting is finished, the results of the previous meetings are not that interesting anymore, as work has progressed. Here we will summarise the highlights of the Riga delegates meeting.!

Riga June 2015The 41th e-IRG Delegates meeting took place in Riga on June 2nd 2015. In the same week a number of e-Infrastructure related events took place in the Latvian capital, including the e-IRG workshop, the Baltic HPC and Cloud Computing Conference and the WIRE Conference. The focus of this closed

delegates meeting was on the next work programme of the European Commission's H2020 and the collaboration with ESFRI. The delegates also decided to start working on a new e-IRG Roadmap which should replace the previous one that dates from 2012. The e-IRG delegates are representatives of the European countries that provide advice on e-Infrastructures to the countries, the European Commission and all organisations involved or interested in e-Infrastructures.

Just before the meeting, the European Commission (EC) had finished the evaluation of project proposals for a call under the Work Programme 14/15. The EC will fund 34 projects including 8 Centres of Excellence, 8 Virtual Research Environments and 6 data & computing projects. Large infrastructures like GEANT, PRACE, OpenAIRE, and EGI will also receive funding for their next phase. Support projects that get funded, include OpenMinTed and RDA.

Work Programme 2016/2017The next two-year Work Programme for 2016/2017 will be published in September. The total budget for e-Infrastructures will be 221 M€ for the two years. In June the European Commission was still actively collecting comments.

Many comments were received on the draft Work Programme. The main change compared with the previous Work Programme is the new architecture that includes integration and consolidation of platforms, prototyping platforms, and support policy as separate topics. International cooperation, the Human Brain Project and the Square Kilometre Array are also expected to be included.

In the Directorate Research and Innovation of the European Commission (DG RTD) the 15 proposals for support of ESFRI projects from the last call have been evaluated favourably.

The structure in the next Work Programme 2016/2017 for DG RTD is similar to that in the previous one. One Call (100M€) will support pan-European Research Infrastructures and has 4 actions: design studies for new Research Infrastructures (20M€); support of new ESFRI projects (40M€); opportunities for other Research Infrastructures that are not part of ESFRI; and the European Science Cloud. In addition there will be calls for supporting projects.

e-Infrastructures also have the attention of the highest political level in the European Union: The European Council. On May 29, the Council made several observations in this regard: The Council "STRESSES the importance of PRACE, a world-class European High Performance Computing (HPC) infrastructure for research that provides access to computing resources and services for large-scale scientific and engineering applications; ACKNOWLEDGES the need to develop the new generation of HPC technologies and CALLS for the reinforcement of the interconnected network of data processing facilities GEANT. In this respect, INVITES ESFRI to explore mechanisms for better coordination of Member States' investment."

e-IRG is closely working together with ESFRI on supporting large Research Infrastructures. One activity will be to draft a document that analyses the European e-Infrastructures landscape. It will be based on the e-IRG document "Best Practices for the use of e-Infrastructures by large-scale research infrastructures", published earlier this year.

E S F R I R e s e a r c h I n f r a s t r u c t u r e s

O t h e r R e s e a r c h I n f r a s t r u c t u r e s

International Research Projects

T h e e - I n f r a s t r u c t u r e C o m m o n s

D a t a

C o m p u t i n g

N e t w o r k s / C o n n e c t i v i t y

T o o l s & S e r v i c e s

E - I R G N E W S L E T T E R

Fall 2015 5

Research infrastructures generate a lot of data. How the management and long term preservation of these data could be handled will be investigated by e-IRG. Of course, e-IRG will make use of the expertise of the Research Data Alliance and experts from the countries that are working already for several years on this topic.

The European Commission organised a big ICT 2015 event in October in Lisbon focusing on European R&D. ESFRI and e-IRG together organised a session at this event.

One of the most important documents of the e-IRG is the Roadmap. e-IRG Roadmaps provide recommendations and visions on how e-Infrastructures should and could be developed over the coming years. The previous version dates back from 2012. A lot has changed. Meanwhile a number of large e-Infrastructures including GEANT, EGI, and PRACE have written new strategy documents. The e-Infrastructure Commons is now well established.

Road map preparation A new version of the Roadmap will be prepared covering:

• New models in research, innovation & learning to cover services related to data infrastructures, virtual environments and multi-domain scientific workflows.

• Easy access to and efficient use of knowledge, implemented in a European Science Knowledge Base for exchanging information between disciplines and projects.

• Virtual Research Environments (VRE) with integrated environment of visualization, computation services, data storage, measurement devices.

• Scientific workflows.• Opening e-Infrastructures for massive data

collection using new (mobile) technologies including the Internet-of-Things, Crowd sensors and Machine-to-machine technologies.

Of course, the e-IRG Roadmap will concentrate on consequences on strategy and policy that may have an impact on e-Infrastructures, such as privacy and governance. A first draft of the Roadmap is expected later this year.

The next e-IRG delegates meeting will be held in Luxembourg on September 17th. For more information about or comments on the

e-IRG work you can always contact an e-IRG delegate in your country. !

Further readingFull public minutes http://e-irg.eu/documents/10920/261473/Public+summary+41th+e-IRG+delegates+meeting.pdf

Reports on the Riga eventVideos of the workshop presentations https://vimeo.com/eirg/videos

Baltic HPC & Cloud Computing Conference 2015 https://www.youtube.com/playlist?list=PL3ODjj4AihxP7-s-yD08yFYVr9yLOs6il

Open Science perspective highlighted at e-IRG workshop in Riga http://e-irg.eu/news-blog/-/blogs/open-science-perspective-highlighted-at-e-irg-workshop-in-riga

The importance of being earnest about the long tail of research https://www.youtube.com/watch?v=NG7MydN44JMRelated documentse-IRG document section http://e-irg.eu/publicationsOther relevant documents in the e-IRG Knowledge Base http://knowledgebase.e-irg.eu/policy-and-strategy

Open Science perspective highlighted at e-IRG workshop in RigaIlmars Slaidins gave a short summary of the Open Science perspective presented at the e-IRG workshop, held June 3, 2015 in Riga at the Baltic HPC & Cloud Computing Conference 2015 on June 4, 2015 in Riga. At the e-IRG workshop the policy problems related to e-Infrastructures were addressed: how to supply Open Science initiatives with support measures from e-Infrastructures?

One of such initiatives is related to the e-Infrastructure Commons. The aim is to integrate all the e-Infrastructure services so that there is no separate network, HPC nor software services but it has to be all integrated and provided for researchers.

How can we open up science, starting from Open Data, Open Software, Open Infrastructure? There are many political issues, according to Ilmars Slaidins. For example, if we have a computer in one country and people from abroad or from other organisations would like to use it, how can we

deal with the financing of this activity, with access authorisation and so on. There are a lot of problems to solve.This is the reason why the European Commission is proposing different political initiatives: how can we open up science to support the development of science? This means that if one scientist has done some research, he might be willing to share his data with others. The other scientists simply continue the research on the same day, doing additional analyses and getting new results.

One of the points is that maybe citation will not only happen in papers but could also be applied to data sets. This is not yet recognized in the scientific community, Ilmars Slaidins explained. When it will come to the situation that data sets could be equally cited as papers, people will be more open to share data sets. There are still many problems though. This was discussed at the e-IRG workshop.

There were not only guests from the European countries but there were also people from the United States of America, including the general secretary of the Research Data Alliance, Mark Parsons, with a very good presentation. There were participants from Canada, Russia and South African Republic. Colin Wright was from the South African Republic and Ilmars Slaidins talked to him. He said that he was really happy to attend the workshop and to listen to the presenters from the European and other countries because in South Africa, they experience exactly the same problems. It was

interesting that he was recognizing that they have the same problems as everywhere else around the world.

The e-IRG workshop partners really hope that it will go on. The workshop was very successful. The e-IRG Group which is directly related to the development of e-Infrastructure policies was very satisfied with the fact that new people joined the workshop from the part of the HPC and other infrastructure segments. The e-IRG partners saw an opportunity for a good exchange by communicating with and involving more people.It is not so easy to share contacts between the computing people and the people from other science disciplines and to link them. We have to work all together. This is the issue about the e-Infrastructure Commons. Now, we have TERENA, the European network organisation, which has joined with DANTE in the GEANT Association. Hopefully, other organisations as well will follow. Maybe in the future, we will have some kind of e-Infrastructure community organisation, Ilmars Slaidins concluded. !

Ilmars Slaidins was the co-chair of e-IRG, responsible for the local organisation of the e-IRG workshop in Riga.

E - I R G N E W S L E T T E R

6 Fall 2015

At ISC 2015 in Frankfurt, Germany, we talked with Augusto Burgueño Arjona from the European Commission, European e-Infrastructure Programme, about e-Infrastructures. e-Infrastructures refer to research and innovation infrastructures and to ICT infrastructures, so basically networking, computing and data infrastructures. The European Commission funds the research projects which covers: networking, GEANT; high performance computing, EGI and PRACE; and projects for data like EUDAT, open science and open access. Those are the major projects. In addition, there are also the projects supporting the Centres of Excellence for HPC, virtual research environments to cover certain thematic areas. Basically, the EC's objective is to fund the ICT layer of e-Infrastructures that are needed for the researchers to carry out research.

You are funding the e-Infrastructures themselves and the development but there are also parts of the Commission that develop the real research too?

Augusto Burgueño Arjona: Historically, DG Connect has been focusing on the ICT part and now, the colleagues of DG RTD focus more on thematic infrastructures. Since these thematic infrastructures are becoming more data-driven, what we are now beginning to observe is that there are certain infrastructures, like ELIXIR, for example, which is very much focused on ICT and ICT infrastructures, and the real challenge is to make sure how we rationalize funding and services so that we focus funding for e-infrastructures in a way that services are also offered to the thematic infrastructures. In this way, we find an appropriate balance of what's devoted to e-infrastructures, the ICT layer, and what is devoted to the thematic infrastructures which are more focused on the specific needs of certain scientific communities.

How does the European Commission decide on what to fund? There is the Horizon 2020 Programme which decides on the projects?

Augusto Burgueño Arjona: We have a two-year programme cycle. Every two years, we

produce a work programme which is the result of consultation with our beneficiaries but also with the scientific community in general. It takes into account what we have been funding in the past, of course. It goes through an evaluation process by the member states through the programme committee that supervises our work in the research programme.

The next period will be 2016-2017. When will that be officially announced?

Augusto Burgueño Arjona: We are in the last steps of the process. The work programme will be hopefully adopted in the month of October. Then it will be made public. What we expect, is that, at the ICT Conference of October 20-22 in Lisbon, we will be able to announce it and explain it in detail to the potential beneficiaries.

Are there already general things known about the programme?

Augusto Burgueño Arjona: The part we are responsible for, the e-Infrastructures part, has been fully discussed and certain elements have already been presented. We are in the final stage - we cannot distribute the document, of course - but we can already explain the main changes that we are introducing. In the past, we used to have a work programme architecture which was very much based on technological layers. We were funding independently network infrastructures, distributed computing, data, and so forth. What we observe is that there is a convergence that is happening, not only from the perspective of technology but also from the perspective of operators and

service providers. The differences between these layers have become more and more blurred. We are shifting the architecture now to a service-oriented mechanism. The focus will now be on services and we will fund services.

In order to do that, we have organized a work programme in two parts: one that we call integration. What we expect is that the current operators of networking, distributed computing, and so forth, will make the necessary efforts so that the services that they provide are integrated and can be seamlessly combined. That is one part. Then, we have another part that we call innovation where we hope to encourage the development of new services. The basic idea here is that we acknowledge the know how of our current operators and for that, we have reserved some budget for what we call platform-driven innovation. That budget will be devoted for the current operators to keep on developing new services on their

platforms. But we are putting extra focus on what we call user-driven innovation. The objective is to encourage new actors, SMEs or innovative companies to take advantage of services that already exist, either for the development of their own business, or to develop new additional services on top of the existing infrastructures.

Our objective is to foster and encourage a really open ecosystem of e-Infrastructure services that in the long run will be exploited in all manners possible and really open to imaginative and innovative ideas.

In the community there are people, for instance in e-IRG, that are talking about an e-Infrastructure Commons and in EGI where they are talking about an Open Science Commons. Is that related to this?

Augusto Burgueño Arjona: I have just described how we will organize the funding. We definitely support the vision of the Commons. We also strongly support the concept of the European Research and Science Cloud that has been recently put forward as well. We think that these things are complementary. We are really concerned

European Commission's Work Programme 2016-2017 is focusing on funding e-Infrastructure servicesInterviews with Augusto Burgueño Arjona

E - I R G N E W S L E T T E R

Fall 2015 7

about how the European budget is allocated for e-Infrastructures and we believe the most efficient way is to fund services. Whether this results in a natural structuring or a natural way of implementing the Commons, or a natural way of implementing the European Research and Science Cloud, has to be seen. From my perspective there are high chances that if we are successful in the Calls with good proposals and the vast definition of services, the natural consequence will be, at least at a European level, that we will be able to identify which those Commons are and we will be able to identify which are those services that the European Research and Science Cloud offers.

We think that it is consistent, that it is compatible, that it enforces those conceptualizations, so to speak, of the e-Infrastructure but these are, as I said before, measures about how to structure funding.

When we understand correctly, the way to structure funding of e-infrastructures is more focused on services and they are building blocks. When the community says we should better organize the building blocks in such a way that they are a kind of Commons then that is something that the community should do.

Augusto Burgueño Arjona: That is correct. What we also hope, as you well know, is that the e-infrastructures is not just a European effort. As a matter of fact, our effort in terms of budget, is limited compared to the effort that the member states are putting in the development of infrastructures. The member states are confronted with the same type of problems. For historical reasons, funding has been organized per layer, per technological layer. This might show the way perhaps to an alternative way of funding which might encourage member states as well to consider structuring their funding in terms of services in which case you will really have a very natural way of defining the Commons and a very natural way of identifying which parts of the Commons are funded by whom. We will see, we don't know, we don't have a magic ball but our main concern, as I said, was to make sure that the services that are funded at a European level, are well identified, well documented, fundable, and so forth.

More details will be seen and we will explain, like for example the catalogue of services that we want to encourage to be developed. We are taking very serious steps towards having rock-solid KPIs, cost models, and the rest. We really are expecting to have a solid package that will eventually facilitate these developments of the Commons and the European Research and Science Cloud. !

http://ec.europa.eu/digital-agenda/en/e-infrastructures

European super-computer programme in next Horizon 2020 period programme to also focus on HPC access for SMEs

At ISC 2015 in Frankfurt, Germany, we had the opportunity to talk with Augusto Burgueño Arjona from the European Commission, Unit of e-Infrastructures, including European Supercomputer Service Programme, who also delivered the welcome address at the Conference stating that "HPC is high in the European political agenda as it is recognised as an essential technology to respond to the growing needs of our modern societies, transforming societal challenges into business opportunities, and bringing growth and jobs to the European economy". Augusto Burgueño Arjona also stressed that "there are important upcoming steps, such as the large-scale platform integration of the technology building blocks and the coordinated acquisition at EU level of world-class systems".

What is the role of HPC in Europe and Europe's role in the world?

Augusto Burgueño Arjona: HPC has been a priority for Europe since a few years already. As you know, we have a European HPC strategy which the European Commission is supporting, based on three basic pillars: technology development towards exascale; building capacity and access to HPC systems; and the involvement of user communities so that they can really benefit from all the possibilities that HPC offers. Part of what I said in my welcome address is that we have achieved a lot so far. There are very successful projects. Some of them even are in the third generation of funding. Now, it is time to start thinking about how to take stock, how to integrate all these technologies in new prototypes, in new demonstrators. This is basically what we are looking at now in the preparation for the next steps.

What are the next steps? Will you be focusing on research or also on industry?

Augusto Burgueño Arjona: In addition to the demonstrators that I just mentioned, there are two other additional elements. One is the coordination of procurement policies in the member states. We really believe that coordination will be needed given the high cost of these computers and the short life cycles. We have already sensed that some member states would be willing at least to explore that possibility. The other element which is as important, if not more important, is facilitating access to HPC resources for SMEs and other innovative actors. This is in the context of the Digital Single Market strategy that has just been adopted by the European Commission.

The basic objective and the key headings are the digital access of industry which means to facilitate, no matter which sector we are talking about, that companies fully exploit all the possibilities of ICT - and this includes of course HPC technologies. It also involves the

HPC Ecosystem

EXDCI Eurolab-4-HPC

ExCAPE

GreenFLASH

READEX ESCAPE

INTERTWINE ALLScale

ANTAREX

ExaFLOW

ComPat

NLAFET ExaHYPE

NEXTGenIO SAGE

ExaNEST

ExaNoDe ECOSCALE

EXTRA Mont-Blanc 3

Mango

Results of 2014-2015 Calls

Project Start: Autumn 2015

BioExcel COEGSS EoCoE E-CAM ESiWACE MAX NOMAD PoP

Centres of Excellence

Key EU developments HPC

Results of the 2014–2015 FET HPC calls.

E - I R G N E W S L E T T E R

8 Fall 2015

democratization, if you want, of access to HPC because our understanding is that there are a number of industries, and a number of actors within those industries, that have understood and have developed the mechanisms to really make use of HPC technologies. There is a huge amount of companies, some of them which are also providers of services and products to these big players that today don't have access. So we really have to explore mechanisms, so that HPC technologies are made accessible and available also to minor players or important players but without the financial muscle to really invest in HPC technologies.

There is already a longer history of smaller scale, you could say, and assisting SMEs with HPC technologies, including the UberCloud which in fact also is coming from Europe, and the Fortissimo and CloudSME projects. So, what you want to do is to bring this to a bigger scale?

Augusto Burgueño Arjona: In the next programme period in 2016-2017 we are going to try to experiment with a cooperation model which will also involve innovation clusters, as we call them, regional clusters that have a mission to develop their

local economies by fostering the development of new SMEs, by making resources available to them. What we want to explore is whether bringing together those local entities which mission is to help SMEs in bringing them together with infrastructure operators like HPC providers, like PRACE, for example, could facilitate their access because today the mechanisms that we have, are only European in the sense that if you are a local SME and you want to use a PRACE machine, you have to know that PRACE exists and you have to know that they have a programme called SHAPE for which you have to apply. We believe that SMEs naturally are known by their local organisations and those are better placed to help. So that is something that we are going to prototype in the next programme period.

Then, we will also be exploring options to increase capacity, work mechanisms of coordinating local and national budgets that could be used so that regions that are specifically specialized in certain areas, could involve, for example, a structural funds for the development and for the deployment of HPC capacity in a way that could not only be beneficial for their local companies but also at the European level if this is in a coordinated

manner. For that, we are also relying on political initiatives like the Juncker plan, for example, which is a mechanism to facilitate funding and loans to companies which really has the potential to create synergies between all the funding mechanisms like Horizon 2020 and structural funds, for example.

Juncker is the president of the European Commission. What is his plan about?

Augusto Burgueño Arjona: The Juncker plan is a mechanism to facilitate the access to the European Investment Bank (EIB) loans. So, basically, the European Commission, together with the European Investment Bank, have created the guarantee that enables the European Investment Bank to take higher risk projects. So, one could envisage, for example, a project to deploy HPC capacity in a few member states with the participation of private partners like telco providers or even users, like financial companies or energy companies. At the same time, the funding of these private entities that will provide for the deployment of this infrastructure could be complemented with public funding, coming from other sources like, as I said, the structural funds or Horizon 2020. !

European Commission has launched Digital4Science PlatformThe European Commission has launched the Digital4Science platform set-up to stimulate conversations on Excellence in Science and its activities around Open Science, e-infrastructures, Future & Emerging Technologies and the FET flagships.

With this platform, the European Commission wants to give stakeholders a voice: to talk to them, listen to them, and engage in an on-going dialogue on the EC's activities and policies.

To participate, you must register and create a profile.

The European Commission has launched 5 discussions ("Ideas") focusing on e-infrastructures topics with the objective of fostering cooperation among projects in specific fields:

• Open Scholarship

• Skills & Training

• Biomed & life sciences

• Earth Science, energy & environment

• Social Sciences and Humanities

Further discussion streams on issues the EC considers more relevant, such as KPIs, TRLs and catalogue of services will be launched at a later stage. The input will feed into the e-concertation meeting that the European Commission will organise in Brussels on the

9th of November 2015 for invited projects. It will be a great opportunity to take stock of what has been discussed on Digital4Science and complement it with face-to-face conversations. !

The platform is available at: http://ec.europa.eu/d4science.

E - I R G N E W S L E T T E R

Fall 2015 9

Around EuropeCountry newsCountry description in the e-IRG Knowledge Base updated

The country description in the e-IRG Knowledge Base has been actualized and updated with new information on the national infrastructure, policy and documents. http://knowledgebase.e-irg.eu/countries

GreeceSuccessful training courses organized on Greece's first supercomputer

In September 2015, the Greek Research and Technology Network (GRNET) has organized a series of successful training sessions on ARIS, the first national high performance supercomputing system supporting large-scale scientific applications. The supercomputer was installed at the end of 2014.

The supply and installation of the system was handled by COSMOS Business Systems SA, in collaboration with IBM, after an open international tender conducted by GRNET. The system plays an important role in the development and promotion of scientific research in Greece and in southeastern Europe. It is integrated into the European "ecosystem" of supercomputers, and included in the 500 most powerful computers in the world. The system will strengthen the participation of Greece in the pan-European infrastructure PRACE.

The national supercomputer was developed within the project "PRACE-GR Supercomputing Services for the Greek Research and Academic Community" which is funded by the Operational Programme "Attica" and the European Regional Development Fund (ERDF).

The system will strengthen the Greek scientific community, covering the needs of Greek users in multiple scientific sectors. Computational chemistry, physics, biology, biomedicine, meteorology, seismology, computational engineering and materials

science are just some scientific areas that largely rely on the use of modern supercomputing infrastructures.

The system is based on the NeXtScale platform of IBM and incorporates the latest generation of Intel Xeon E5 v2 processors with Ivy Bridge technology, providing a computing power that reaches 180 TFlop/s.https://hpc.grnet.gr

PolandUniversity of Warsaw to publish report on Open Science in Poland 2014

In August 2015 the Interdisciplinary Centre for Mathematical and Computational Modelling at the University of Warsaw, issued a report on Open Science in Poland in 2014. This report was prepared as part of the Open Science Platform project. The report was edited by Jakub Szprot. Contributors are Andrzej Lesniak, Michael Morys-Twarowski, Krzysztof Siewicz, Michal Starczewski, Lidia Stepinska-Ustasiak, and Jakub Szprot.

The most important element of open science in Poland in 2014 is Open Access to scientific content, while other elements play only a minor part. This also applies - in particular - to research data whose role in open science becomes more and more prominent in Europe and all over the world. Open Access is mostly implemented by publishers of scientific journals, researchers themselves, institutions providing IT solutions and few institutions managing repositories. Almost half of all Polish scientific journals are Open Access journals. There is a growing popularity of Open Access practices in the scientific community, and the beliefs on the advantages of openness are taking root. There is a growing and developing Open Access e-infrastructure. Open repositories are established, although still only on a small scale. These phenomena provide good ground for further implementation of open models, whereas the main challenge is the implementation of solutions - legal, technical, organisational and financial - that would allow to use the full potential of openly available

content. This mostly applies to expanding the repository infrastructure and developing optimal use of the infrastructure, allowing for unified access to diverse resources on a national level.

A serious problem, both in the context of international scholarly communication and the European Commission policy, is that Poland lacks institutional Open Access strategies and policies, both on the government level, and the level of research-funding and scientific institutions. International experience shows that without such strategies and policies an efficient implementation of Open Access is not possible. At the same time, it appears that due to the shape of Polish science system, developing, adopting and implementing such strategies and policies could help Poland to quickly catch up to international standards. http://knowledgebase.e-irg.eu/documents/243153/292773/Open+Science+in+Poland+2014

Short takesIn the Netherlands a new permanent committee will draft a roadmap for research infrastructures

The Swedish roadmap to infrastructures has been published in English and on the web. There is a strong focus on e-Infrastructures. Next year the Swedish government will release the next bill for innovation and science.

Germany has published a call on the process for a roadmap and for Research Infrastructures. The deadline is 15 January 2016.

Estonia has updated its roadmap and this is the basis for funding of infrastructures. It is available on the web in Estonian and English. An investment plan for infrastructures will be available at the end of this year.

Norway has awarded in June about 200 million euro for e-infrastructure. About 20 million euro is directly for e-infrastructure with a large focus on open data and sharing data. The update of the roadmap will be ready end of the year. !

E - I R G N E W S L E T T E R

10 Fall 2015

IDC assessment of European HPC strategy Remarkable progress, but more needs to be done to reach exascaleIDC has delivered a report to the European Commission about the HPC developments in Europe. It contains an assessment of the HPC Action Plan Objectives set out a few years ago. The report also provides some recommendations. The report is now at the European Commission, who will study it, draw conclusions from it and use it in further development of the HPC actions. In general, IDC concludes Europe has made good progress in implementing HPC and exascale plans. Europe is now on par with other major HPC players in the world. In the recommendations IDC advices Europe to expand funding for HPC, concentrate on the whole software stack, and also look at the newly developing High Performance Data Analysis (HPDA) area.

According to the IDC report, Europe has made impressive progress in areas that are crucial for the goals of the HPC Action Plan (High Performance Computing (HPC): Europe's Place in a Global Race (February 2012)), especially organizing the European HPC community to pursue HPC leadership on a unified basis, expanding the scientific and industrial access to and use of supercomputers, and launching initiatives to strengthen the European HPC supply chain.

The IDC assessment of progress in implementing the European HPC strategy, and the related recommendations that IDC made in earlier versions of the report 3 years ago, are based on the four Action Plan Objectives and the six related Strategic Actions. The ultimate aim is to make Europe stronger in using HPC to advance science and industry.

According to IDC, Europe has significantly narrowed the wide gap separating the most capable U.S., Chinese and Japanese supercomputers from their European counterparts. In November 2010, 9 of the world's 50 most powerful supercomputers

were located in Europe. Four years later, Europe hosted 19 of the top 50, including the PRACE tier-0 supercomputers.

According to an IDC study on European projects, European HPC investments are producing excellent returns-on-investment (ROI) for science and industry. IDC captured detailed ROI information on 143 European HPC projects. For projects that generated financial returns, IDC sees each euro invested in HPC on average returned 867 euro in increased revenue/income and 69 euro in profits.

IDC concludes Europe achieved healthy HPC funding growth in 2010, 2011 and 2012, but Europe-wide funding declined heavily in 2013 and 2014. On average, however, the net funding increase over the last five-year period was extremely good for pursuing HPC leadership. Significant investments will be needed, however, for pre-exascale systems in the period 2019-2020 and exascale systems in 2022, says IDC. IDC estimates that the overall European-wide HPC R&D

investments, supercomputer purchases, and related services in 2010 amounted to approximately 1,8 billion euro: 715 million euro in purchases of supercomputers, plus 90 million euro (in supporting investments, integration services and professional services); plus about 850 million euro (storage, software and repair services purchased with the computers); plus about 30 million euro (from the EC); and about 90 million euro (R&D by vendors and users). IDC says these are IDC estimates, as there is no European tracking of these numbers at this time.

The main actors - the European Commission, PRACE and ETP4HPC, have done an admirable job of coordinating with each other to advance the European HPC strategy, but more needs to be done, says IDC. To amass

the 1 billion-plus euro in funding needed to acquire pre-exascale and exascale supercomputers in globally competitive time frames, the European Member States will need to find a way to pool resources and the EC will need to find a way to contribute a significant portion of the funding.

IDC believes HPC procurements in Europe should make greater use of PPI mechanisms, and ETP4HPC and Member States within PRACE should coordinate closely to ensure that the ETP4HPC roadmap reflects the innovations PRACE members and others would like to procure.

At the very high end of the supercomputer segment, Europe has significantly narrowed the former wide gap separating the most capable U.S. and Japanese supercomputers from their European counterparts. Spending increased substantially in the EU/EU+ for large supercomputers from 2009 to 2012, says IDC, but then declined. European overall spending on large supercomputers grew from 112 million euro in 2009, to 658 million in 2012, then down to 362 million euro in 2014. In November 2010, shortly after the founding of PRACE, 9 of the world's 50 most powerful supercomputers were located in Europe. Four years later, in November 2014, Europe hosted 19 of the top 50, including the PRACE tier-0 supercomputers. The aggregate peak performance of the Europe-based supercomputers rose more than ten-fold during this period, from 4,3 petaflops in November 2010 to 48,9 petaflops four years later.

IDC sees that from a technical standpoint, Europe's HPC community, building on existing and planned EU- wide HPC development initiatives, is well positioned to exploit a strong base of indigenous and foreign technologies across its commercial, academic and government sectors to assemble exascale HPC capability that could, in some critical application sectors, achieve world-class stature or even global leadership. A clearer path is needed, however, for driving innovation into supercomputer procurements and pooling enough money to deploy pre-exascale and exascale supercomputers in globally competitive time frames.

Progress on the Four EU HPC Action Plan ObjectivesThe overall European HPC strategy is built on a foundation of key objectives that are the source and inspiration for specific strategy actions. IDC has examined progress against these key objectives in detail, and after lengthy discussions with many EU and other HPC technology and policy experts, offers the following assessment.

Digital Agenda for Europe

High Performance Computing in the EU: Progress on the Implementation of the

European HPC Strategy

Final Report

A study prepared for the European Commission

DG Communications Networks, Content & Technology

E - I R G N E W S L E T T E R

Fall 2015 11

Assessment of Specific Action Plan Objectives1. Provide a world-class European HPC infrastructure, benefitting a broad range of academic and industry users, and especially SMEs, including a workforce well trained in HPC.

IDC assessment: Strong progress was made in deploying large HPC systems across Europe, primarily funded by the Member States, with important secondary funding support by the European Commission through PRACE. Although the overall level of funding was considerably lower than recommended in IDC's 2010 report (as shown in Section 5, Table 5.G), Europe nevertheless made outstanding progress in getting Europe "back in the pack" of top global research areas over the last four years.

2. Ensure independent access to HPC technologies, systems and services

IDC assessment: Hardware technologies from European suppliers today have a very small share and presence in HPC across Europe. Some European software companies are highly successful in Europe and across the world. ETP4HPC collaborations between European and non-European suppliers will benefit European scientists and engineers. A large majority of the European HPC stakeholders IDC interviewed agreed that European scientists and engineers require access to best-in-class HPC technologies and systems, no matter where in the world they come from.

3. Establish a pan-European HPC governance scheme to pool enlarged resources and increase efficiency, including through the strategic use of joint and pre-commercial procurement

IDC assessment: Pre-commercial procurement (PCP) and public procurement of innovative solutions (PPI) are underutilized in Europe today. PCP and PPI are key mechanisms used by the governments of the United States, China and Japan (Europe's main rivals for HPC leadership) to drive commercial competitive advantage and to advance suppliers of indigenous technologies. European HPC stakeholders greatly prefer PPI (and related mechanisms) over PCP as a way to advance innovation. There is no

central procurement agent in Europe with the financial ability and motivation to exploit PPI or related mechanisms on behalf of Europe, as needed to compete successfully with the U.S., China and Japan.

4. Ensure the EU's position as a global actor

IDC assessment: Advances in highly parallel software will be one of the most important determinants of future global HPC leadership. Europe has world-class strengths in highly parallel software. The de-emphasizing of funding for exascale software development by the U.S. government has created an opportunity for Europe to gain an important advantage. EESI's exceptional work has set the stage for this. The cPPP on HPC, established with the ETP4HPC with 700 million euro in EC funding, provides a framework for making it happen, especially through the Centres of Excellence and continued EESI work.

Assessment of the progress and success of the six Strategy ActionsBased on the overarching objectives defined in the European HPC plan, specific policy-related strategy actions were created. For those six strategy actions, IDC offers the following further assessment and recommendations.

1) Governance at EU level: seeks adequacy, openness, and efficiency of the current organisations (e.g., PRACE and ETP4HPC) to structure the industrial and scientific stakeholders, to steer the high level objectives and policies on HPC, to pool available HPC resources across the Member States, and to efficiently implement the HPC strategy.

IDC assessment: The key organisations contributing to the implementation of the European HPC strategy - the European Commission, PRACE and EPT4HPC - have done an admirable job of advancing Europe's position in the few short years since the 2012 Communication. No single European Member State has the financial and related means to compete effectively with the U.S., China and Japan for HPC leadership. If Europe is to be an HPC leader, it will therefore be necessary for Member States to coordinate their HPC strategies more closely, including the pooling of funding, and for the EC and the Member States to coordinate even more closely. This tighter coordination will require some adjustments to existing governance rules and practices by all parties. Tensions already present within Europe's loosely coupled, collaborative HPC governance model will likely grow as the exascale computing era approaches, unless these issues are addressed. Governments of some Member

States hosting PRACE tier-0 supercomputers are reluctant to continue funding 100% of the substantial, rising operating expenses associated with these computers. They would like some relief from the Commission and/or from other PRACE members. Access to more of the tier-0 supercomputers' cycles may need to be secured for European, as opposed to national, use.

In carrying out this study, IDC said it found that even some of the most prominent members of Europe's HPC community did not adequately understand the impact of

governance models on the European HPC strategy - or how all the pieces of the strategy fit together. Stronger communications outreach is needed to convey this understanding.

2) Financial envelope for HPC spans current investment levels for the acquisition of high-end HPC resources in Europe, and analysis of required levels to meet the Action Plan objectives (including investments for system acquisition, training, HPC software and applications, etc.).

IDC assessment: The European Commission's HPC investment levels have grown substantially, through a range of initiatives including the 700 million euro, multi-year investment to support the future-oriented cPPP being carried out with the ETP4HPC. But Commission contributions have fallen short of the amounts recommended in IDC's

E - I R G N E W S L E T T E R

12 Fall 2015

2010 report. The Commission has contributed to support PRACE, and plans to start helping to support some procurements for large supercomputers in 2016.

3) The implementation of funding mechanisms, such as pre-commercial procurement in the public sector (the major buyer of high-end HPC) and pooling of research resources, to support HPC suppliers for developing a leadership-class HPC system about every 2 years.

IDC assessment: According to IDC, PPI and related mechanisms are crucial for attaining and maintaining HPC leadership. Europe's rivals for HPC leadership - the U.S., China and Japan - regularly employ these mechanisms. These mechanisms are used today only occasionally in Europe. No clear strategy exists for pooling resources to finance the acquisition of pre-exascale supercomputers.

4) Development of European state-of-the-art supply capabilities needed for European independent access to key HPC technologies, systems, services and tools for Europe (including level of pre-commercial procurement and other R&D investments, support to European HPC suppliers, jobs in European HPC supply industry, etc.).

5) Industrial exploitation of HPC including regional/national centers for the access of industry (including SMEs) to HPC (HPC Competence Centers), industrial HPC-based development and innovation, education and training in HPC, HPC trained workforce in Europe, and more.

Europe already has some of the world's leading HPC centers for collaborations with industry, including SMEs. Many of the European HPC stakeholders IDC interviewed for this study agreed that European programmes supporting industrial access and collaboration, such as PRACE, SHAPE and Fortissimo, have been successful but need to do more. Only a small percentage of European SMEs that could be helped by HPC seem aware of these opportunities, for example. The EC hasn't done an optimal job of communicating the HPC strategy, resources, plans and contact persons to the broader HPC community, and in particular to industry.

6) Ensuring a level playing field, in particular regarding inequalities in HPC market access and exploitation obligations regarding intellectual property rights of HPC results generated in Horizon 2020.

IDC assessment: Europe has long been the world's most open HPC market. Government HPC markets in the U.S., Japan and China all present barriers to non-domestic HPC suppliers, although the private-sector markets in these countries are more open and both government and private-sector markets are

generally open to non-domestic commercial software.

Overall RecommendationsBased on an examination of the European HPC objectives and related strategy actions, IDC offers the following recommendations, each keyed to one or more of the strategy actions.

Recommendations on the Strategic ActionsIDC makes the following targeted recommendations:

1) Expand Funding for HPC (Financial Envelope & Funding Mechanism, and Needed for Making Europe World Class in HPC). PRACE members and the EC should agree to provide significant funding support to acquire two pre-exascale supercomputers in 2019-2020 and two additional exascale supercomputers in 2022. One path should stress European pre-commercial technology. The total recommended HPC cumulative funding increase (for all parties) from 2016 to 2020 is just over 1 billion euro. The European Commission should extend the end date of the Action Plan from 2020 to 2022, to match the expected exascale time frames of the U.S., China and Japan and make it easier to amass the funding levels recommended in this study. The Member States and the EC may need to adapt their practices in order to pool the money needed to fund exascale systems.

2) Improve Communication of the Strategy (Supply Ecosystem and Needed for Making Europe World Class in HPC). The European Commission should create a single website portal enabling access to comprehensive information on the European HPC strategy. This needs to include a single person in charge and as a direct contact for information and for answering questions. This is key for getting more users, ISVs and SMEs involved.

3) Develop the HPC Ecosystem (European Supply Capabilities and Governance). ETP4HPC should continue to support collaborations involving European suppliers and (often much larger) non-European suppliers. This will accelerate the learning curve of some European suppliers who are less experienced. European suppliers will benefit from competing on equal terms with suppliers not based in Europe. Software will be one of the main determinants of future HPC leadership and Europe is very strong in parallel software development..

4) Improve Support for Industry (Industrial Use of HPC). PRACE should consider promoting SME and industry adoption of HPC with a SHAPE initiative that lets SMEs

and other industrial firms that are new to HPC try it out without cost, and without needing to show scientific merit. Europe already has some of the world's leading HPC centers for collaborations with industry, including SMEs. Centers with strong industrial experience are well positioned to mentor centers with less experience with industry.

5) Improve Skills and Talent (To Ensure Europe as a Global Actor and Needed for Making Europe World Class in HPC). The European Commission should undertake a communications campaign to update the image of HPC as a career choice.

Additional Recommendations from IDCHigh Performance Data Analysis (HPDA):This is the market for Big Data workloads that are complex or time-critical enough to require HPC resources. No dramatic shift in the European HPC strategy is needed for HPDA.

Cloud Computing: In IDC's worldwide studies of high performance computing (HPC) end-user sites, the proportion of sites employing Cloud computing - public or private - has steadily grown from 13.8% in 2011, to 23.5% in 2013, to 34.1% in 2015. As with HPDA, IDC believes that no major change to Europe's HPC strategy is needed to accommodate the growing importance and technical evolution of public and private Cloud computing.

Co-Design: The EU should initiate an EU-wide exascale test bed programme to provide strong support for exascale co-design capabilities.

Centers-of-Excellence: These fall under the cPPP on HPC. The centers should focus heavily on software - not just applications but also the whole HPC software stack. At a minimum, one center should focus on the software requirements of industry, and one center be devoted to HPDA.

ReferenceHigh Performance Computing in the EU: Progress on the Implementation of the European HPC Strategy, Final Report, A study prepared for the European Commission DG Communications Networks, Content & Technology by IDC. ISBN number 978-92-79-49475-8 !

http://knowledgebase.e-irg.eu/documents/243153/246094/High+Performance+Computing+in+the+EU+-+Progress+on+the+Implementation+of+the+European+HPC+Strategy.pdf

E - I R G N E W S L E T T E R

Fall 2015 13

IntroductionAround 80 persons attended the e-IRG Workshop in Riga, which was organised during the Latvian EU presidency of the European Union.

The workshop addressed the Policy Aspects of Open Science for e-Infrastructures. The workshop objective was to discuss the transition to Open Science, with an emphasis on the implications for e-Infrastructure policies.

The topics discussed included the policies on governing the access to data; computing and networking resources; sharing of resources and the transfer of knowledge; the perspectives of the Latvian EU presidency; the EC Consultation on Science 2.0 results; and the policy implications for e-infrastructure when Open Science is introduced.

E-IRG chair Sverker Holmgren opened the workshop and talked about the idea of e-Infrastructure Commons as a one-stop shop and an innovating system for new services. This has been widely accepted, Holmgren said, but further work is required in order to define this concept.

Next speaker was Lauma Sika, Counsellor, Attaché (Research and Space questions), Permanent Representation of the Republic of Latvia to the European Union. Lauma Sika introduced the priorities of the EU under the Latvian presidency: having set an agenda for a competitive, digital and engaged Europe, along with the news from the European Competitiveness Council. The Council held a

policy debate on open and excellent science, as a follow-up to the Science 2.0 public consultation carried out by the Commission.

One of the key outcomes was the idea of a European Open Science agenda. Europe is in the process of creating a Digital Single Market, and the development of a European Open Science Programme is also one of the priorities. The EU Council of Competitiveness invites ESFRI to explore mechanisms for better coordination. Open Access could further increase the use of public funding. In addition, science metrics need to be reflected upon including an upcoming revision of the EU copyright law. Investment in open access research data intensive research should also be improved. In a recent meeting, the EU Competitiveness Council invited ESFRI to explore mechanisms for better coordination of Member States investments in e-Infrastructures, Lauma Sika said.

Jean-Claude Burgelman, European Commission, DG RTD presented a keynote: “EC Consultation on Science 2.0 results”. He noted that there is a paradigm shift in the ecosystem of research infrastructures and research data. The Science 2.0 public consultation launched by the EC resulted, amongst other things, in a name change. The term that has been accepted is Open Science, Burgelman said. New ways of measuring the impact of Open Science are required. Most respondents recognise the trends included in the consultation paper. Availability of digital technologies and their increased capacities has been identified as the most powerful drive of this change, followed by ‘Researchers looking for new ways of disseminating their output.’

Burgelman summarized the conclusions of the consultation. European Open Science Agenda. Potential actions under consideration:

Fostering Open Science: Creating incentives and removing barriers, e.g.• Establish a stakeholders forum at European

Level and a self-regulation/ clearinghouse mechanism for addressing Open Science issues

• Propose a European "code of conduct" setting out the general principles and requirements of how Open Science should affect the roles, responsibilities and entitlements of researchers and of their employers

Mainstream Open Access to publications and data, e.g.• Consider extending the Horizon 2020 pilot

on Open Access to data

• Develop EU guidelines for addressing IPR issues and the funding of data-management

Introducing Open Science actions to address common societal challenges under the European Research Area and under Horizon 2020

Develop data infrastructures for Open Science, e.g.• Mandate the development of common

interfaces and data standards• Coordinate at European Level the funding,

maintenance and interoperability of research infrastructures

• Support the development of a European Open Science Cloud for data, protocols and methodologies

The European Commission will also support the development of a European Open

e-IRG Workshop RigaOpen Science & e-Infrastructures

June 3, 2015

Marcin Ostasz, Fotis Karayannis, e-IRGSP4

E - I R G N E W S L E T T E R

14 Fall 2015

Science Cloud for research. This will provide all EU researchers a virtual environment with free, open and seamless services for data storage, management, analysis and re-use, across disciplines. The cloud will federate existing and emerging horizontal and thematic data infrastructures, effectively bridging today’s fragmentation and ad-hoc solutions

Anders Flodström, EIT ICT Labs, tackled the topic of “Education & Innovation & Jobs” - Still Silos’ in his keynote “Open Innovation and Open Science - Brainchildren”.

‘Higher Order Skills’ are needed in order to secure a harmonised growth. We experience an explosion of communication and data – big business, big data, big science and also a disruption of our industries in the form Industry 4.0 (Cyber-Physical production), Internet of Things, 3D “printing”, clean technology, renewables, etc. Today’s technical progress is a lot faster than social and cultural advances. The new paradigm involves the move from a producer to a consumer economy, from efficacy to innovation, and from quality to dream. Innovation is becoming the leading factor in the economy, generating 75% of the economic growth. There is a huge lack of ICT skills in Europe, while most European economies are experiencing high unemployment rates. Innovation creates and destroys jobs. Job change cycles are much faster than educational cycles. Re-education and multiple competences are required now; jobs change faster than people can change. Universities must become a stakeholder among others in the new knowledge society, said Flodström. This is a shift from a Knowledge Society to a Skill & Competence Society.

Flodström introduced the new concept of Knowledge Triangle which includes Research, Education and Innovation as the key factors adding value to the economy, society and science. Massive Open Online Education (MOOC) could be a mechanism to address the skills gap. Examples of the job areas in demand in the new World are: A Digital World – the Cloud and DNA, a World of Seamless and Ubiquitous Communication and Internet of Things. In the coming ten years millions of new jobs will be created and millions of old jobs will disappear, while jobs will change faster than education.

Session 1: Open Science - Directions and main issuesThe objective of this session was to present main directions on Open Science along with main issues or areas that may need policy actions.

Sami Niinimaki of the Ministry of Education and Culture, Finland, explained the Finnish

Open Science and Research Initiative 2014–2017.

Finland has an updated roadmap for Open Science and RI, mapped onto the ESFRI roadmap, which is included in its Open Science and Research 2014-2017 strategy. This initiative sets to increase the quality and competitiveness of Finland’s research and innovation system. By increasing openness in research, we will simultaneously be improving reliability, transparency, and the impact of research. Openness also creates opportunities to participate in scientific advancement, and enables easier and more effective utilisation of research results. Promoting open science and research requires not only extensive involvement from the research community, but also cooperation and coordination, internalising new ways of working, and developments in research environments, researcher services and research infrastructures.

Its main recommendations are: 1. Improvements of access to and collaborative use of research infrastructure and 2. Shoring up of the funding base of research infrastructures.

In an Open Science maturity assessment, basic information from openly available material on the web has been collected to assess the open science operating culture.

Wolfram Horstmann, University of Goettingen, Germany and LIBER presented “Beyond Big Bata — The Long Tail of Research”. Universities and research institutions are identifying research data as an asset and invest in collecting and providing access to those datasets generated at their institution that do not fall within the scope of other discipline-based, or government repositories. This "long-tail" (e.g. FigShare, My Experiment, ‘50% of research is done on data sets that are less than 1 GByte') can pose challenges for those institutions, including the diversity of disciplines, the variety of standards, the multitude of projects and persons involved.

Yet, Horstmann said, the support of the long-tail is paramount - not in the least because the highest innovation capacity in research is in the emerging fields - not in the already established mainstream. The challenges are data quality, discoverability (with 50% of data in the university lab and 39% on the university server) incentives, and creating a feasible business case and improving access to datasets. RDA has established a Long Tail of Research Data Interest Group to tackle this issue. The main conclusions in this area are: simplification to Big Data – otherwise, Europe’s potential will not be utilised - data diversity, and the institutional perspective needs to be fostered.

Mark Parsons, Rensselaer Polytechnic Institute, USA and RDA Secretary General presented “Data Policy for Open Science”. Open science requires open data. Data policy should serve the interests of the organisations sponsoring the data collection. With research data the sponsor is often governments so the interests should be those of the public, and it is the public that ultimately benefits from open science. Mark Parsons recalls that the vision of RDA is to encourage researchers and innovators to openly share data across technologies, disciplines, and countries to address the grand challenges of society. As its mission, RDA builds the social and technical bridges that enable open sharing of data. RDA serves as a Policy Test Bed – it is not a policy organisation, but it can help implement policy.

Preservation requirements are well defined in the Open Archive Information System (OAIS) Reference Model, Parsons said, but there is no similar model for access requirements (not even a common definition of “access” and what restricts it). Some areas (e.g. biomedical) require unique access. The key recommendations are: Mind your preservation and access - your stewardship, Clarify and credit roles, Promote and empower the champions - those who add generative value, Look for consensus and emergent norms from the data science community, Iterate.

Session 2: Open Science - Use CasesThe objective of the second session was to present some concrete use cases on Open Science, including best practices, challenges and plans. New ESFRI/Research Infrastructure cluster projects and e-Infrastructure data projects were introduced focusing on their approach regarding Open Science. The issue of open software licenses was also tackled.

Mark Allen, CDS Strasbourg, France described in his presentation the “Open Science in the framework of the ASTERICS Astronomy ESFRI cluster”. Astronomical research is facing the challenges of combining data from many

The e-Infrastructure Commons needs to be part

of Europe’s strategy to create the Digital Single

Market (DSM). The impact of Open Science on e-Infrastructure and the

role of e-Infrastructure in this process should be

assessed in order to enable that inclusion.

E - I R G N E W S L E T T E R

Fall 2015 15

different telescopes. The Horizon 2020 'Astronomy ESFRI & Research Infrastructure Cluster’ (ASTERICS) addresses the cross-cutting synergies and common challenges shared by the various Astronomy ESFRI facilities (SKA, CTA, KM3Net & E-ELT). ASTERICS will enable astronomers to have broad access to the data products of the ESFRI telescopes via a seamless interface to the Virtual Observatory framework, embracing the Open Science concepts.

According to Mark Allen, the ‘Virtual Observatory’ will include archives and databases form a ‘digital sky’. It will open new possibilities via data discovery, efficient data access and interoperability, driven by exploding data rates as well as multi-wavelength, time domain & survey science.

Allen sees as benefits of being open and interoperable:

• Transition in the way Astronomy is done

• Opening up the research process

• Access, Interoperability

• Engagement – scientists, data providers, citizens

The astronomical research community is willing to lead the way with the biggest infrastructures as participants in defining the Virtual Organisation framework.

Stuart Lewis, Edinburgh University Library, UK presented “Research Library Support for Open Science - supporting the transition”. With the move towards open science, research libraries can play a key role in aspects such as open access publishing, supporting research data management, and providing access to digital collections of information, Lewis noted.

The Library has a role in coordinating infrastructure and services that have been made available to every researcher at the University of Edinburgh. It assists with undertaking well-managed research, having been driven by local University and national policies towards open science. This includes a suite of services and tools to create data management plans, store and sync data, archive it safely, publish it openly, and share it

with the world. In particular, these tasks include: Facilitating Open Access publishing, Support Research Data Management, Providing access to digital collections.

Natalia Manola, University of Athens/ATHENA Research Center, Greece talked about a new project: “Text mining: the next data frontier. An infrastructural approach”.

Recent years witness an upsurge in the quantities of digital research data, offering new insights and opportunities for improved understanding. Text and data mining is emerging as a powerful tool for harnessing the power of structured and unstructured content and data, by analysing them at multiple levels and in several dimensions to discover hidden and new knowledge. However, text mining solutions are not easy to discover and use, nor are they easily combinable by end users.

Manola continued about the new project OpenMinTeD that aspires to enable the creation of an infrastructure that fosters and facilitates the discovery and use of text mining technologies and interoperable services. It examines several use cases identified by experts from different scientific areas, ranging from generic scholarly communication to literature related to life sciences, food and agriculture, and social sciences and humanities. OpenMinTeD text mining tools, services and associated resources will run on the cloud, requiring an in-depth optimization of service deployment and execution via scalable VM-based service distribution and use of distributed storage. One of the needs identified is the need to share infrastructure cost.

Neil Chue Hong, Software Sustainability Institute, UK discussed “Open software for (open) science”. Modern research relies on software. However, it can be difficult to understand the implications of the way in which software is licensed.

Open source is now de facto for infrastructure software. Open source encourages Exploitation, Reproducibility and Robustness, Reuse. Open source helps support open science, Hong said. Open Source should fulfill the following conditions:

Non-discriminatory - Research should not be restricted or siloed

Access to source code - Research should be transparent, robust, and accessible

Redistribution of software - Providing access to the widest possible community

Removing barriers to reuse - Research should encourage building on the work of others, and giving them credit

Session 3: Panel discussion: Open Science The policy perspective and relevance to e-InfrastructuresThe objective of the Panel discussion was to present Open Science policy perspectives at all levels and discuss about the relevance and implications for e-Infrastructures . The panel was chaired by Françoise Genova, CDS, France. Joining her were the panelists: Jean-Claude Burgelman EC DG-RTD; David Bohmert, ESFRI representative, Swisscore; Bob Day, Janet and JISC; Iveta Gudakovska, Latvian representative in OpenAIRE project, Library University of Latvia; and Sami Niinimaki, Ministry of Education and Culture, Finland.

Norbert Meyer, PSNC, asked the panel: How can we convince people to be involved in Open Science? David Bohmert answered: This is what we have been doing in ESFRI since 2000. Now the Competitiveness Council has asked us in its conclusions to assess the investments strategies in e-Infrastructures and the coordination among Member States. A confederated approach is very promising.

Jean-Claude Burgelman noted the principle of reciprocity applies also in the case of the Open Science Cloud: be part of the integrated cloud only if you are contributing. Bob Day added that there is the subsidiarity principle also, which says that only the tasks that cannot be performed at the local/national level, should be performed at the EU

E - I R G N E W S L E T T E R

16 Fall 2015

level. So, when it comes to the users we have to check first at national level and sometimes at the local level. There are barriers in the universities and research centres. They should allow new models such as the cloud to be used by their users.

Sverker Holmgren referred to the presentation by Wolfram Horstmann on the long tail of science, questioning that there is no Nobel prize coming out of big data; yet there is a Nobel prize in software. Still, it is not only about developing the software but also maintaining that software (which is very challenging). Bob Day noted that to capture these assets, such as software, is just one part of the problem; we also need to understand them: discover, understand what they are supposed to do and what they are doing and manage all this.

Jean-Clause Burgelman said that if all the fragmented needs of the long tail of science are aggregated, then they will become visible.

According to Dimitrios Koureas, the technical challenges are not so important. We have e-infrastructures. Do we have users that are using these e-infrastructures? Yes! Do we have pioneering communities? Yes! Do we have users not using them? Yes! Can we afford going ahead without them? No! Two ways of involving them are the following: One way is an institutional way and the other way is a more thematic way (domains/disciplines). We do not talk enough about these and we need to fund enough of these.

David Bohmert concluded that we need to go ahead beyond institutes and domains towards education and a knowledge society.

Tiziana Ferari said EGI mentioned that there are 7 ESFRI Research Infrastructures talking about their own clouds. It is important to understand how these initiatives are funded at the national level. This may be an action for e-IRG, she said. In this way confusion will be avoided. Maybe it is important to create an e-IRG working group about the national level funding and the coordination with other levels.

Panel chair Françoise Genova summarized the panel discussion in a few sentences. The

e-Infrastructures need to engage all types of users including individual researchers, communities, citizens, and even kids from schools. This means that e-Infrastructures

need to be built with the data user needs in mind. In this way they will become relevant and usable. e-Infrastructures should gradually become transparent to the users being able to use them for their research. Sustainability of e-Infrastructures is key for the users to invest in them and use them.

Research e-Infrastructures should be combined with services from industry, in first instance to serve demanding users and guarantee sustainability, and in second instance to serve commodity needs and the long tail. The distributed and confederated approach as well as the related business and funding models need to be studied. The education of children is of the highest relevance because education if the future for everything. Children have to be involved in e-Infrastructures. Data and software stewardship also is crucial. This is not only about developing software and storing data, but also about maintaining them in order to be fully usable. !

The full workshop minutes, slides and video recordings of the presentations can be found on the e-IRGS website:http://e-irg.eu/e-irg-workshop-june-2015

WIRE 2015 Summary of the Session on Open DataGeorgia Tzenou, National Documentation Centre, Enterprise Europe Network-Hellas

The "Week of Innovative Regions - WIRE 2015" that was held in Riga, Latvia, June 4-5, 2015, included a session on the efficient use of Open Data for regional planning and growth.

The main questions addressed were: How can Open Data be used for regional development?; Which priorities are relevant to e-infrastructure for boosting transparency, performance and engagement till 2020?; Are there any successful examples of using Open Data for the generation of new public services?

The conclusions of the second Plenary session on "The challenge of Open Data" were the following:

• Research data openness to the current Latvian scientific community is a huge challenge.

Workshop conclusions

The funding of Open Access to scientific publications, research

data, software, methods, educational material and other resources needs to be addressed, in particular the provision of

support for access and storage/maintenance

New ways of measuring the impact of Open Science are

required.

The creation of the European Open Science Cloud for Research

needs to be addressed (as an incarnation of the e-

Infrastructure Commons).

A framework to be able to share infrastructure costs is essential.

There needs to be a close working relationship with ESFRI to

address the coordination process of Member States investment

strategies in e-Infrastructures.

A joint ESFRI-e-IRG Working Group needs to be established.

The e-Infrastructure Commons has another dimension that is

Global, EU, regional, institutional, etc.

E - I R G N E W S L E T T E R

Fall 2015 17

• Scientific data are important to perform all sorts of activities. There is a lot of work involved when data becomes scientific. The discussion about data has made good progress in Nordic countries.

• The NordForsk is like a body that works with common resources. There is a Nordic funding available and there is also funding attracted from different countries.

• The Arctic region is a good testing area. It might have great opportunities, huge possibilities, and a strong interest for commercial activities. They have identified areas that are of interest for the future development of the Nordic region where ICT tools can be used for future development.

• The aim of the industry is to transform science by using tools, networks and media, and make research more global and close to society.

• The researchers play a very important role. Scientists want to perform research of high quality and for that they need access to the best research tools. Industry needs to be competitive. Apart from that there is a need to provide best services.

• It is time start working now, to make all this happen.

• It is important to have a view from different perspectives.

• If you can quickly change something you have more money to invest in the infrastructure.

• We have to keep focusing on the effect of Open Data but just providing a lot of open data is not a goal in itself. !

PLAN-E Workshop reportRosette Vandenbroucke, VUB

The PLAN-E open workshop and closed meeting were organised on September 3-4, 2015 in Munich, Germany and hosted by Dieter Kranzlmüller, in conjunction with the 11th IEEE International Conference on eScience 2015.

The workshop was opened by Dieter Kranzlmüller and a PLAN-E introduction by Patrick Aerts. Anton Frank presented “eScience in Germany” with an overview of the activities and centres involved. The term eScience itself is hardly used in Germany, but hidden under various labels there is a lot of activity that would qualify for eScience, according PLAN-E ‘s Terms of Reference. “Challenges of providing a common format for up-to-date eScience graduate education at Swedish universities and HPC centres” was presented by Anders Hast.

The graduate school combines efforts that entail different aspects of eScience and is planning to include research in less traditionally involved domains, such as humanities and social sciences. Rob van Nieuwpoort spoke about “eSTEP, an eScience Technology Platform”.

This platform is much more than a repository of software proven to be practical, suitable and maintainable. It also encompasses the research required to keep the knowledge level of eScientists up to date. It is open to users from elsewhere.The first keynote speech was delivered by Wilco Hazeleger: “To get science from eScience”. Wilco Hazeleger showed various examples of high level science efforts clearly enabled by the use of modern techniques from ICT and the potential of e-infrastructures.

The second keynote was given by Bob Bishop: “Engaging the Public in Science Narrative - the use of modelling, simulation, visualisation and big data tools”. Bob Bishop demonstrated, through many practical examples, the close interaction of the various, important, global and special dynamic

processes that currently are individually simulated and analysed, but actually would require a step towards their interactive coherence.

The two working sessions were moderated by Neil Chue Hong. The first one dealt with the question of how PLAN-E can be instrumental in knowledge exchange and exchange of other eScience and data science related information. The second one tackled the issue of how PLAN-E can be instrumental to get a solid policy for software sustainability and data stewardship established for academia in Europe. Small groups discussed the issues and Neil Chue Hong summarised the group results.

The second day of the workshop was only open to Plan-E members. Four members presented eScience efforts in their country: Austria, Turkey, Belgium and France.

There were three external presentations. The first one by Yin Chen from EGI addressed EGI moving toward an Open Science Commons. The second presentation was offered by Neil Chue Hong from SSI who talked about the Software Sustainability Institute that is informing national policy. The third presentation was held by Colin Wright from South Africa. He highlighted eScience and eScience research in South Africa and gave his view on the next steps to take.

Discussions reflecting on these presentations concerned the position of parties relevant to the eScience scene, such as SSI, EGI and international organisations. PLAN-E intends to work at the European level in order that international organisations will have a point of reference if they want to engage with European eScience matters. At the same time, PLAN-E will pro-actively seek international contacts to exchange information, broaden the scope and vision on eScience and to engage actively with international parties. It was agreed that until further notice or changes in vision or structure of PLAN-E, SSI and a representative from South Africa will have an observer status. EGI and JISC already had this status.

During the second meeting of PLAN-E, it was decided that a White Paper will be compiled on eScience and Data Research. The work on this document continues. !

E - I R G N E W S L E T T E R

18 Fall 2015

ETP4HPC, EXDCI and SESAME Net New HPC initiatives in Europe's Horizon 2020Ad Emmen, July 2015, Frankfurt

In the Horizon 2020 session at ISC 2015 in Frankfurt, Germany, Chair Panagiotis Tsarchopoulos from the European Commission introduced four speakers: Jean-François Lavignon from Bull, Sergi Girona from PRACE, François Bodin from IRISA, and Karen Padmore from HPC Wales. But first, Panagiotis Tsarchopoulos started out by sketching the key HPC developments in the European Union. In 2012, a report has been issued titled "HPC: Europe's place in a global race".

Afterwards, the Council on HPC in 2013 concluded in a document to the "Establishment of a European Technology Platform on HPC and a Strategic Research Agenda". This has led to the HPC chapter in the Horizon 2020 programme including three elements: Computer science evolving towards exascale HPC; achieving excellence in HPC applications; and providing access to best supercomputing facilities and services for both industry and academia via PRACE.

The three elements are heavily interrelated, explained Panagiotis Tsarchopoulos. Specifications of exascale prototypes have to be developed and there has to be decided on technological options for future systems. The EC believes in a strong collaboration between HPC centres and Centres of Excellence for applications. In addition, it is important to identify applications for co-design.

The Call results for exascale technologies can be divided in two parts. The first one addresses HPC core technologies; the second one involves HPC ecosystem development by means of coordination support actions. These do not involve research. The first round of calls was closed on 25 November 2014.

There were 81 eligible proposals received that included 652 participants.

The start for the first projects is scheduled in Autumn 2015. Two ecosystem actions have been identified, namely EXDCI and Eurolab-4-HPC.

The HPC Centres of Excellence Call amounts to 14 million euro. The goal is to establish a limited number of user-centred Centres of Excellence. This call was closed on 14 January 2015. The European Commission received 20 eligible proposals. These projects will also start in Autumn 2015. The application areas involve high-energy physics, earth sciences, etc.

The Future Calls are scheduled for 2016-2017. They will address co-design of HPC systems and applications; the transition to exascale computing; and exascale ecosystem development.

The Pan-European HPC infrastructure and services will require an investment of 15

million euro; the Public Procurement of innovative HPC systems has been estimated at 26 million; and the Human Brain Project will receive 5 million euro.

The timeline looks as follows. In 2014, attention was paid to technology building blocks. I 2016, first level co-design will be addressed. In 2017, the exascale transition is planned and in 2018, the second level co-design will follow.

During the ICT 2015 Conference, to be held between 20 and 22 October in Lisbon, Portugal, sessions will be organized about EU HPC funding.

ETP4HPCJean-François Lavignon gave an update on ETP4HPC SRA. This is an open association, that serves as a counterpart to the European Commission, to discuss the HPC policy in

Europe. It has been established in 2012 as a contractual public-private partnership.

The ETP4HPC association has developed a strategic agenda in 2013 and now, an update is happening. At present, there are over 170 experts involved in 8 working groups. Workshops are being organized, such as EESI2, and there are also workshops for end users and ISVs. There are 20 people represented in each working group.

Jean-François Lavignon explained that there are four axes, including:

1. the HPC stack elements consisting of the HPC system architecture, system software and management, and the programming environment

2. extreme scale requirements: balance compute subsystem, I/O and storage performance and improvement of system and environment characteristics

3.new HPC deployment in the areas of mathematics and algorithms as well as HPC usage models

4.HPC usage expansion: usability, affordability, HPC services; focus on SMEs; and education and training

On 25 February 2015, kickoff meeting sessions have been launched. In March 2015, workgroups have been started. In May 2015, there was a Levelset with EESI.

EXDCI initiative Serge Girona talked about the EXDCI project. Partners in this project are PRACE and ETP4HPC. The project has a duration of 30 months, starting September 2015. The budget is 2,5 million euro.

The objectives are to coordinate the development and implementation of a common strategy for a European HPC Ecosystem.

The strategic goals are to develop a common European HPC strategy and to operate a synchronised European HPC community for joint community structuring and synchronisation.

The partners will draw up an overall planning and work plan structure with periodic updates of the Strategic Research Agenda and

E - I R G N E W S L E T T E R

Fall 2015 19

they will also update the PRACE scientific case.

Work packages include:

• WP2: technological ecosystem and roadmap toward extreme and pervasive data and computing

• WP3: applications roadmap toward exascale

• WP4: transversal and strategic prospective

• WP6: international liaison

• WP7: impact monitoring: methods and tools

The next actions and events are the KoM Meeting, September 2-3 in Brussels, Belgium; the European HPC Summit, May 9-13, 2016 in Prague, Czech Republic; the coordination initial workshop, September 29-30 in Rome, Italy, organized by EXDCI. The participants will be PRACE, ETP4HPC, CoE, and FETHPC projects partners. They will discuss with each other to understand the whole ecosystem.

François Bodin addressed the technical and scientific challenges in the EXDCI initiative. He emphasized that HPC is strategic for economical growth and that it is a key enabler for the high tech industry.

The HPC challenges are paths to exascale while considering parallelism, fault tolerance, energy efficiency, and heterogeneity. The different Laws, including Moore, Dennard on constant energy and Kryder on storage density, are reaching their limits.

Energy consumption is dominated by data movements. There are also other challenges including the memory wall, the data explosion, the diminishing MTBF, and economical constraints.

The EXDCI initiative is interdisciplinary CSA. There are two scientific actions, namely the ETP4HPC and PRACE scientific case. There will be Eurolab4HPC and EXDCI collaboration on the roadmap, as well as on training and innovation.There is SRA themes

coverage through the new FETHPC RIA.

François Bodin told the audience that energy and programming environments are dominating the landscape but that the data are under-represented.

There is a worldwide initiative to share and build an international plan for developing next generation open source software for scientific HPC. This is called the International Exascale Software Initiative.

The synergies for the ecosystem are increasing but the economical dimension has to be taken into account. The EUrolab4HPC/EXDCI collaboration also has a strong worldwide dimension with regard to the presence of a Big Data and Extreme-scale Computing representative.

SESAME NetKaren Padmore represented the SESAME Net initiative and expanded on this Supercomputing Expertise for Small and Medium Sized Enterprises Network initiative.

The support of one network of HPC competence centres will promote access to computational expertise anywhere in Europe and enable the dissemination of best practices in the HPC industrial landscape, she explained.

The aims of SESAME Net are coordination, outreach, training and exchange of best practices but there will also be actions carried out in other programmes.

The centres will be set up specifically to promote access to industry and SMEs for economic development and regeneration purposes. The centres now have expertise but little industry engagement.

The SESAME Net kickoff meeting was held on 9-10 July, 2015. The partners will now be setting up the network to create a widening membership. The collating materials are already available. The partners also intend to organize workshops and publish a calendar of events. !

EGI to receive new statutes and new nameThe EGI Council has decided to simplify the name of the European Grid Infrastructure to EGI - just the acronym without an expansion. The change reflects the evolution of EGI towards becoming a worldwide, technology-agnostic e-Infrastructure for science.

The change of European Grid Infrastructure to EGI is part of an update in the statutes of the EGI Foundation signed over the summer. The new statutes also approved important changes in the governance of the EGI Foundation:

• Full membership is now open to non-European countries: participation is no longer restricted to a geographical area.

• EIROs, ERICs and other organisations that support the objectives of EGI can also become full members from now on.

"The changes are meant to emphasize the international interest and mission of EGI, which are not limited by a particular technology or geographical area", stated Yannick Legré, Managing Director of the EGI Foundation. "We aim to serve excellent science, regardless of location and discipline."

Matthew Dovey, Chair of the EGI Council, added: "Updating the statutes of the

foundation and the EGI name will reinforce cooperation with other pan-European and worldwide e-infrastructures and research infrastructures, bringing us closer to the way science is actually undertaken." !

Source: EGI

E - I R G N E W S L E T T E R

20 Fall 2015

Data is Light – observations from the 6th RDA Plenary Anni Jakobsson - CSC, Finland

"It is like opening the curtains - the more you open the more is unveiled. Data is light." The words from French Minister of State for Digital Technology Axelle Lemaire, who spoke at the opening session of the 6th RDA (Research Data Alliance) Plenary. The meeting took place on 23-25 September in Paris, France.

The Plenary attracted a record audience: over 700 participants took part to the meeting. In total RDA has over 3200 members from 103 countries. It is a remarkable figure since RDA was just launched in 2013. RDA is working on building the social and technical bridges to facilitate data sharing and re-use on a global scale. "We are the ones who are building the bridges", stated RDA Secretary General Mark Parsons.

Keynote speaker Barbara Ryan, Secretariat Director at Group on Earth Observations aptly put it: "Countries have borders, research doesn't". This is exactly the vision for RDA: researchers and innovators openly sharing data across technologies, disciplines, and countries to address the grand challenges of society.

RDA currently has 56 Working and Interest groups, and the three days were packed with Working and Interest group meetings as well as 26 Birds of a Feather (BoFs) meetings where new ideas and potential groups are discussed. In these groups, experts come together to exchange knowledge, share discoveries, discuss barriers and potential solutions, explore and define policies and test as well as harmonise standards to enhance and facilitate global data sharing.

Coupled with this RDA boasts a broad, committed membership of individuals and organisations dedicated to improving data

exchange. Four new deliverables from the Working and Interest groups were published at the 6th Plenary meeting bringing the total to 8 with 4 new ones from the RDA - WDS (World Data System) due early next year. The outputs are summarized in the Outputs Booklet.

A new phase of RDAAdoption of the RDA results is essential. This is the way the highly valuable outputs produced by the RDA Working Groups are taken into practice.

"We have moved into a new phase in RDA. I'm pleased to see how much adoption there is", Parsons continued.

The adoption of the RDA outputs had taken a leap forward from the 5th Plenary in San Diego half a year ago. This time seven adopters from EU and the US presented how they have taken RDA recommendations into use. The adopters from the US presented at the Plenary were: Deep Carbon

Observatory, Platform for Experimental Collaborative Ethnography, Datafed.net and the Materials Innovation Infrastructure.

First European adopters of the RDA results were presented too. The European RDA forerunners are EUDAT Collaborative Data Infrastructure, German Climate Computing Center (DKRZ) and Common Language Resources and Technology Infrastructure (CLARIN).

These institutions had taken into use e.g. the RDA Data Type Registry (DTR), which helps researchers to visualise the contents of unknown data types, or the RDA Data Foundation & Terminology (DFT), which helps the cross disciplinary data exchange and interoperability between scientific communities.

Enterprise engagementThe theme of this Plenary was enterprise engagement with a focus on research data for

climate change. Twenty-one companies with solutions or products with a focus on climate change were showcased during the Experimentation Day. The goal of the Experimentation Day was to foster exchange between RDA members and data related company representatives, sharing views, challenges and dreams about data sharing.

Making environmental tracking personalPlume Labs was one of the companies showcased during the day. The company is using open data to raise the awareness about air quality, and predict the air quality conditions. Using their apps, users in polluted cities are better aware when, for example, to go running, take kids to the park or eat outside and this way decrease their environmental health risk by avoiding excessive exposure at peak times.

The company's aim is to develop a dense net of personal air quality sensors, and this way

users can get more detailed information about the air quality in the exact spot where the user is. The data collected by these portable sensors is also crowdsourced to create a better coverage of air quality sensors throughout the world even to places where air quality sensors wouldn't otherwise exist.

"Air pollution kills more than AIDS and malaria combined. By breathing polluted air, you lose life expectancy. For example if you live in Paris, you

have 6 months less life expectancy due to air pollutants. There are 7 million avoidable deaths by just following the quality of air that we breathe", said Romain Lacombe, CEO of Plume Labs.

Plume Labs also was one of the winners of the RDA Climate Change Data Challenge and they will present their solutions at the COP21 conference in Paris in December 2015 together with the two other Climate Change Data Challenge winners Vizonomy and BioVeL.

Computing processors heating waterOne of the companies showcased during the RDA Experimentation Day was Defab. The company is developing green cloud computing via innovative technology. The company has developed a water heater which is combined with computing processors. The

E - I R G N E W S L E T T E R

Fall 2015 21

European exascale projectsDEEP-ER and Mont-Blanc to investigate new programming and network-attached memory technologies

At ISC 2015 in Frankfurt, Germany, we interviewed Filippo Mantovani from the Mont-Blanc project and Estela Suarez from the DEEP-ER project on the opening night of the ISC 2015 Exhibition at their booth showcasing European exascale projects. Juri Schmidt demonstrated the network-attached memory technology used in the DEEP-ER project. Filippo Mantovani demonstrated the creditcard-sized system-on-chip that is used in the Mont-Blanc project.

Estela Suarez highlighted the goals of the DEEP-ER project. In fact, DEEP and DEEP-ER are both European exascale projects. The team proposes a new architecture, that is different from what we know today. It is the so-called cluster/booster architecture. The team builds a booster which is based on Intel Xeon Phi processors and puts all these processors together into their own network which is called EXTOLL in this case. In this way, the team operates accelerators in an autonomous way. The cluster/booster approach allows the researchers to map the scalability levels of the applications into the hardware. So, you analyze your application, you see where your highly scalable parts go fast, you put them under the booster. The low

scalable parts go to the cluster. In the DEEP-ER project we look more into the I/O and resilience aspects. We are looking into new memory technologies. We are enhancing this standard cluster/booster architecture with new memory technologies.

Filippo Mantovani told that the Mont-Blanc projects have a similar structure. There is a Mont-Blanc I and a Mont-Blanc II project. In Mont-Blanc I the main objective is to exploit embedded and commodity technology. This means commodity technology, like what we have in our smartphones and tablets, is used for doing scientific computation. This is the general idea of both projects.

The main target for Mont-Blanc I is to build a real prototype and the team has actually installed it in the Barcelona Supercomputing Center since February this year. Part of it is present at the Student Cluster Competition at ISC 2015 where six students are doing the

challenges of this competition with this Mont-Blanc part.

From the prototype, the team is learning lessons for the next-generation architecture based on basically ARM chips. It also has several scientific applications that are tested on this new architecture.

In Mont-Blanc II the team decided to explore other topics that were not covered in Mont-Blanc I, for example, the resiliency topic that is

a key challenge for exascale systems. The team also improved its tool support because this 'weird' architecture sometimes also requires support of tools for debugging and performance analysis. So, the team has a full coverage of tools within the project.

You are here on a booth with other exascale projects. How important is collaboration for you with all the other projects, not only in Europe but also internationally? Do you also have collaborations internationally? But let's first say how important collaboration in Europe is for you.

Estela Suarez: Just the fact that we can be here on a booth to show with other projects is already a big help to create collaborations with the other projects. It helps to be more active, more present, and have a higher visibility when there are more people involved. From the scientific point of view, these other projects look at different aspects: they look more into the software parts, or into the applications. For instance, there is a project which is focused on industrial applications. It opens up opportunities for further collaboration. We can identify points where we kind of address the same kind of problems, the same kind of issues but with different approaches. We can also see where things are complementary and where some approaches are more efficient than others.

Filippo Mantovani: The key point is sharing knowledge of what we are doing. We basically all know about the challenges. Having a place like this booth means that you are sharing your way to tackle the challenges. If you are strongly obsessed with your project, sometimes you don't even know which project is more or less going in the same direction. So, these are the kind of events that allow us to share what we are really doing as well as the achievements and results. A key advantage of having a booth together is that it isopening up the results to a larger community. The workshop that we are going to have at the end of this week is also opening up our results to a wider community, to show where Europe is going in this race to

heat from the computing processors is used to warm up the water in the tank. The product combines a 200-300 litre water heater with ten computing processor units. The water in the tank is normal hot water that can be used e.g. in households, hotels etc. The customer buys the water heater, and the company pays the electric bill caused by the processor usage. At the same time, the company sells cloud-computing resources for customers for different purposes.

"There are no cooling costs, and no facility management. Using this technology there isn't need for datacenters. This way, we are able to match scientific research and energy savings",

said Benjamin Laplane, one of the founders of the green start-up.

The final prototype will be running by the end of the year, and there will be 10-15 units installed around Paris during next year.

Defab was selected as one of the RDA Experimentation Day winners together with Plume Labs.

Next time in JapanThe biyearly RDA Plenaries are truly global events with active participants from around the word who are dedicated to improve

research data exchange. The main theme of the 6th Plenary, Enterprise Engagement also brought two new organisational members to RDA, growing the total number to 39. Attracting more enterprise involvement in the data ecosystem is key for future development of the RDA.

The next RDA Plenary will take place in Tokyo, Japan 1-3 March 2016 with the theme Making Data Sharing Work in the era of Open Science. !

https://rd-alliance.org/blogs/data-light-%E2%80%93-observations-6th-rda-plenary.html

E - I R G N E W S L E T T E R

22 Fall 2015

exascale, but also in a fine-grained way to show what each of these projects are really doing everyday and what their real impact is.

Do you also collaborate internationally, with Japan, the US or other countries?

Estela Suarez: We are trying to use these events, like ISC but also the Supercomputing Conference in the USA, to network, to look at other people doing similar or maybe different things in other continents. That is the platform we use to foster collaboration. We also, at least in our case, in our institutes, for instance the Juelich Supercomputing Centre, and the other institutes which are involved in the project, have our own collaborations with different institutions all over the world. We also use all these platforms to disseminate the efforts we are doing and to share experiences the same way as we do in Europe.

Filippo Mantovani: It is important that all the partners are international players in Europe and they all have contacts worldwide. We constantly have contacts all over the world. Events like these or the Supercomputing Conference in the US allow us to open up to this worldwide community. In the case of Mont-Blanc for example, we also have this end user group that is a dissemination tool which we use to explain about the Mont-Blanc prototype. There are companies outside there, not only in Europe, but in the whole world, that want to try these kinds of platforms. We offer these companies access to our platform. We exploit this win-win situation because they have the applications and we collaborate in this sense. Basically, it our work has no borders. That's the kind of things that we do at this kind of events.

DEEP-ER technology

Thanks for the introduction. Let's now look at some of the results.

Juri Schmidt explained a little bit more about the technology behind the DEEP-ER projects.

While the DEEP Booster prototype is already running, the DEEP-ER project will investigate new technologies, not only by extending the current commodity devices but also investigating new research areas. One of these is the network-attached memory.

In the DEEP-ER project we have the high-performance interconnect from our partner EXTOLL. In EXTOLL you can connect nodes in a 3D torus switchless. You have six links on the side where you can create the torus. There is an additional seventh link where we are able to plug additional devices. That is, for

example, another hierarchy of memory. What we investigate is the so-called network-attached memory.

The network-attached memory consists of an FPGA and a 3D stacked D-RAM Hybrid Memory Cube (HMC). We then will be able to plug the network-attached memory directly into the EXTOLL network. We are not just extending the network by an additional level of memory but also we will be able to implement specific functions and the ability to process memory in the FPGA. There is a remote host. You don't need, if you want to do an operation on the attached memory, to transfer the data back and forth through the network, to calculate it and write it back to the HMC.

Instead you send a descriptor, which is very small package , through the EXTOLL network to the network-attached memory and you can perform all the calculations and send the result back. These operations can simply be

global sums, reductions but also backup processing, for example.

Another application for the network-attached memory is check pointing. If your system fails, you need a checkpoint and with the hybrid memory cube which delivers a magnitude more bandwidth than traditional systems, you'll be able to get the restart checkpoint as fast as possible. As with any other memory technology you need a memory controller to talk to the HMC because, if the machine is a rather new technology, you have to develop your own HMC controller.

Let me show you what we have achieved. What you see here is a Xlinx FPGA in the middle and a hybrid memory custom-board, developed on our own. It is built for evaluating and for verifying the operation of our own HMC controller. What is good to know is that the HMC controller is made open source, so if you want to have it, you can download it and you can use it for no cost.

Mont-Blanc technologyFilippo Mantovani will explain a little bit more about the MontBlanc technology

We have the Mont-Blanc prototype actually in real life, so what I have here in my hand is a node card of the Mont-Blanc prototype. It is the size of a credit card. It is a system-on-chip that includes dual four ARM plus a GPU. Everything is embedded. We also have RAM memory, some local storage and some network. Basically, what we have here is the capability of a server but in a very small form factor.

What we do then is to integrate fifteen of these node cards in a blade - I was actually saying Bull blade - because Bull is part of the consortium of the project, as well as ARM and other research centres. We use the Bull technology from here to scale up to two racks of this kind of technology.

As I said before, this prototype is installed at the Barcelona Supercomputing Center, so we are using it and one important point is, in order to investigate the energy-efficiency, is to be able to control the power consumption. We have a power monitoring that is at the level of the node so we can really monitor everything on node-level from one to 1080. So we can have fine-grained power monitoring and we are collecting all the power data, temperature data, everything we can gather at the node level. We are collecting it in a database so this means that we can do a sort of data analysis.

You can submit the job in our cluster and you can do energy data

processing post mortem. You run the job, you finish, you get your job, you have the starting point, and you have the energy to solution. This is a key point: the amount of energy that you are using in an application. In this sense, we can investigate how efficient our different implementations are or how efficient the programming model is that we are introducing and using in this kind of architecture.

Naturally, we also need a programming model. The software model that is being used in the Barcelona Supercomputing Center is called COMPSs and basically is a forerunner of OpenMP 4. It is a task-based programming model that we are using to schedule tasks.This is basically what we have running and what we are investigating everyday. We are really excited because we can really control a lot of parameters. We can measure and having from these measurements a more clear understanding of where this technology is and it can lead us in our investigation. !

8

Exynos 5 compute card 2 x Cortex-A15 @ 1.7GHz 1 x Mali T604 GPU 6.8 + 25.5 GFLOPS 15 Watts 2.1 GFLOPS/W Carrier blade 15 x Compute cards 485 GFLOPS 1 GbE to 10 GbE 300 Watts 1.6 GFLOPS/W Blade chassis 7U 9 x Carrier blade 135 x Compute cards 4.3 TFLOPS 2.7 kWatts 1.6 GFLOPS/W

Rack 8 BullX chassis 72 Compute blades 1080 Compute cards 2160 CPUs 1080 GPUs 4.3 TB of DRAM 17.2 TB of Flash 35 TFLOPS 24 kWatt

Mont-Blanc [GFLOPS/W]

Green500 [GFLOPS/W]

Nov 2011 0.15 2.0 Nov 2014 1.5 5.2

The Mont-Blanc prototype

Frankfurt, 16/07/2015 ISC 2015 - EEP workshops 8 The Mont-Blanc prototype

E - I R G N E W S L E T T E R

Fall 2015 23

The importance of being earnest about the long tail of researchAd Emmen

3 June 2015, Riga

At the e-IRG Workshop, held on June 3, 2015 in Riga, Latvia, we had the opportunity to interview Wolfram Horstmann, Director of the State and University Library at the Georg-August University in Göttingen, Germany, and member of the LIBER Executive Board, which is the European Association of Research Libraries. Wolfram Horstmann held a presentation at the e-IRG Workshop about the long tail of data. There are several aspects that Wolfram Horstmann usually wants to mention about the long tail when he is offered the chance to do so.

The first one is that long tail is not what it seems. It is sometimes thought that the long tail contains smaller and superspecific parts of research but looking at, for example, the distribution of funding, one can see that there are many more projects that receive a smaller amount of budget and very few projects that receive a large amount of funding. There are many more projects that produce smaller kinds of data than projects that actually produce Big Data. The majority of research is about the long tail of data.

The second important aspect about the long tail of research and, correspondingly the long tail of research data, is that once you have

accepted the fact that we are talking about ten thousands of institutions worldwide, hundreds of thousands or millions of projects and definitely millions of researchers, rather than a few that produce a large amount of data, then the challenges' diversity, the diversity of the people, the diversity of the data types, the diversity of regions, countries and institutions and how to manage this diversity should not become an obstacle to what we all want, namely an open and excellent framework for performing research.

FocusRather than focusing on the big infrastructures and focusing on the big projects and Big Data, we actually should focus on the problem of how to overcome that diversity, how to build aggregated environments, aggregated information resources and environments in which researchers can collaborate and build the vision of a globally connected and an open science framework.

Standards and big infrastructures are important, according to Wolfram Horstmann. It is not a contradiction that standards and big infrastructures exist but the point that

Wolfram Horstmann is trying to make is that standards and big infrastructures much more need to take into account the individual, the project's, and the institutional perspective because it is often driven by big projects, big data centres, and the institution or the individual university perspective is sometimes neglected.

AlignmentWith respect to the next phase of the e-Infrastructure development and Europe, what Wolfram Horstmann would like to see is a kind of alignment, closer collaboration and connection between the big infrastructure development and the actual institutional and individual research perspective. While we do excellent research in the area of the big infrastructures, the Big Data initiatives, the mainstream part of excellent research is actually in the long tail of research. !

https://www.youtube.com/watch?v=NG7MydN44JM

Policy actions for “Long-Tail” data

Policy Actions for ‚Long-Tail‘  DataNo. Long-Tail

Characteristic Do’s Don’t’s

1 Heterogeneous Let a 1000 Flowers bloom and embrace diversity

Restrict innovation capacity by prescriptive normalisation

2 Small Provide scale-to-size solutions Enforce Big Data potpourri

3 Uniquestandards

Increase awareness and allow ‚smart‘  standard  customization

Ignore the context-specific expertise of data-management

4 Not regulated Stress open licenses in publishing & career incentives

Limit scientific and economic exploitation or personal rights

5 Individual curation

Support personalization and ownership transfer pipelines

Weaken the responsibility of data producers or researchers

6Institutional, general or no repositories

Support interoperable institutional repositories in a global research data network

Fight distributed approaches and assume  that  only  ‚my‘  data centre

knows how to manage data

http://cdn.zmescience.com/wp-content/uploads/2009/08/jellyfish_nom.jpg

E - I R G N E W S L E T T E R

24 Fall 2015

e-IRG secretariatc/o The Netherlands Organisation for Scientific Research (NWO)P.O. Box 93575NL-2509 AN The HagueThe Netherlands

Phone: +3170 344 0526 Mobile: +316 303 699 04email: secretariat@e-irg.eu

Visiting address:

Java Building, Laan van Nieuw Oost-Indië 300NL-2593 CE The HagueThe Netherlands

e-IRGSP4This newsletter is produced by the e-IRGSP4 project. e-IRGSP4 is partially funded by a grant from the FP7 programme of the European Commission under grant agreement no 632688.

http://e-irgsp4.e-irg.eu

About the e-IRGThe main objective of the e-Infrastructure initiative is to support the creation of a political, technological and administrative framework for an easy and cost-effective shared use of distributed electronic resources across Europe. Particular attention is directed towards computing, storage, and networking.

The e-Infrastructure Reflection Group was founded to define and recommend best practices for the pan-European electronic infrastructure efforts. It consists of official government delegates from all the EU countries. The e-IRG produces white papers, roadmaps and recommendations, and analyses the future foundations of the European Knowledge Society.

http://e-irg.eu

In this issuee-IRG Workshop Luxembourg 2Moving up to an Open Science Commons 3Highlights from the delegates meeting 4Open Science perspective highlighted at e-IRG workshop in Riga 5European Commission's Work Programme 2016-2017 6European Commission’s Digital4Science Platform 8Around Europe - Country news 9IDC assessment of European HPC strategy 10e-IRG Workshop Riga Open Science & e-Infrastructures 13WIRE 2015 session report 16PLAN-E workshop report 17ETP4HPC, EXDCI and SESAME Net 18EGI to receive new statutes and new name 19Data is Light – observations from the 6th RDA Plenary 20European exascale projects 21The importance of being earnest about the long tail of research 23

Upcoming events

3 Nov 2015! Supercomputing for Neuroscientists Workshop, Jülich, Germany! http://www.fz-juelich.de/ias/jsc/EN/Expertise/SimLab/slns/

news_events/2015/scn/_node.html10-13 Nov 2015 !EGI Community Forum, Bari, Italy! https://indico.egi.eu/indico/conferenceDisplay.py?confId=254415-20 Nov 2015! 7th Annual Supercomputing Conference – SC15, Austin, Texas, USA! http://sc15.supercomputing.org/16-17 Nov 2015 – European Data Forum - EDF2015, Luxembourg, Grand Duchy of

Luxembourg! http://2015.data-forum.eu/24-25 Nov 2015 – e-IRG Workshop, Berval, Luxembourg! http://e-irg.eu/luxembourg-20157-8 Dec 2015 – Fifth International Platform on Integrating Arab e-Infrastructure in a

Global Environment - e-AGE 2015, Casablanca, Morocco! http://asrenorg.net/eage2015/