High-Performance Computing in Horizon 2020

Dr Panagiotis Tsarchopoulos

Future and Emerging Technologies

DG CONNECT European Commission

HPC: What for?

Weather, Climate & Earth Sciences New applications e.g. Health, Big data Bio/Life Sciences Fundamental sciences: Physics, Chemistry, Material Sciences, Astrophysics Industrial & Engineering

High Performance Computing

Key EU developments

• Communication from the EC: "High-Performance

Computing: Europe's place in a global race" (2012)

• Council Conclusions on High-Performance Computing

(Competitiveness Council – 2013)

• Establishment of the European Technology Platform on

High-Performance Computing (ETP4HPC - 2013)

Communication on HPC COM "High-Performance Computing: Europe's place in a

global race" (adopted 15 Feb 2012) • • HPC as a crucial asset for the EU's innovation capacity, bringing innovation for societal, scientific and economic needs Ensure European leadership in the supply and use of HPC systems and services by 2020

HPC drivers :

• • • • Addressing Societal/Scientific/Industrial challenges more effectively Industry is increasingly depending on HPC to fulfil the need to innovate in products and services (better jobs) Transition from petascale to exascale computing creates new opportunities for both science applications and computing technologies Gaining independent access to HPC systems and services for Europe supports growth and competitiveness in industry and the economy

Competitiveness Council 29/30 May 2013 Conclusions on HPC Communication

• • • • • HPC is an important asset for the EU's innovation capacity of strategic importance to the EU's industrial and scientific capabilities as well as its citizens: • • • developing innovative industrial products and services, increasing competitiveness, addressing societal and scientific grand challenges more effectively.

All relevant actors, public and private, need to work in partnership; Europe has the technology, knowledge and human skills to develop capabilities covering the whole technological spectrum of the next HPC generation (exascale computing) Importance of developing state-of-the-art HPC technologies, systems, software, applications and services in Europe Invites the EC to elaborate its plans for HPC to support academic and industrial research and innovation under H2020 ;

European Technology Platform for HPC

An industry-led forum founded by stakeholders of HPC technology Through the Strategic Research Agenda, the ETP4HPC has identified research areas and topics to reach a stronger European HPC environment that can benefit Europe and the rest of the world.

www.etp4hpc.eu

The EU framework programme for research and innovation 2014-2020

Excellent Science Industrial leadership Societal challenges

An integrated HPC approach in H2020 "Excellent Science" part of H2020

 HPC strategy combining three elements: (a) Computer Science: towards exascale High Performance Computing; [HPC in FET] (b) achieving excellence in HPC applications; [e-infrastructures] • (c) providing access to the best supercomputing facilities and services for both industry and academia; [e-infrastructures] complemented with training, education and skills development in HPC

Interrelation between the three elements "Excellent Science" part of H2020 Access to best HPC for industry and academia (PRACE)

• • specifications of exascale prototypes technological options for future systems

FET/HPC: EU development of Exascale technologies

• • Collaboration of HPC Centres and application CoEs provision of HPC capabilities and expertise

Excellence in HPC applications (Centres of Excellence)

• • identify applications for co design of exascale systems Innovative methods and algorithms for extreme parallelism of traditional/emerging applications

(a) Exascale technology

• • 

Computer Science: Towards Exascale High Performance Computing

[HPC in FET] The exascale computing frontier requires fundamental science and technology developments to ensure the transition to extreme parallelism and extreme data • evolution of most of the key technological solutions that are satisfactory today will be insufficient to meet the exascale challenge Content: R&D covering the whole spectrum from processors and system architectures to high-level software and tools and novel applications (e.g. encompassing system software, file systems, programming environments and tools, algorithms, mathematics etc.)  engaging a European-wide effort to develop technology to build exascale systems within ~10 years

(b) Achieving excellence in HPC applications



Development, optimization and provisioning of a whole range of strategic applications addressing societal, scientific and industrial challenges to fully exploit the performance of current petaflop and future exaflop systems

[e-infrastructures]  only very few applications using HPC really take advantage of current petaflop systems  codes are developed and maintained on "best effort" approach by individuals for specific communities (and often unclear license terms)  new computational methods and algorithms must be developed, and (new) applications must be (re-)programmed in radically new ways  Content:  establishment of an e-Infrastructure of Centres of Excellence (CoEs) in HPC applications; CoEs focus on scientific or industrial domains that are most important for Europe (multidisciplinary approach)

(c) Access to best HPC for industry and academia



Provide a world-class European HPC capabilities and services to increase competitiveness in science and industry (including SMEs)

[e-infrastructures]  ever growing computational and simulation requirements to stay at the forefront of scientific breakthroughs and to boost innovation capabilities of industry (including SMEs).  access for the best researchers in Europe, independently of the location of either the infrastructure or the user.  Content:  support to Tier-0 HPC capability at the European level  Possible support to procurement through "Public Procurement of Innovative Solutions" (PPI instrument )

Next steps - Horizon 2020

• now: Draft work programmes • now: Discussion with Programme Committees • November 6-8: ICT2013 conference in Vilnius • End 2013: Adoption of the work programme and publication of the calls for proposals • Spring/Autumn 2014: closing of first calls

Indicative timetable – subject to change

ICT in Horizon 2020

Vilnius Lithuania 6-8 November 2013

Conference includes exascale session as well as additional HPC networking and workprogramme sessions

ec.europa.eu/digital-agenda/en/ict-2013 ec.europa.eu/digital-agenda/en/content/action-132-invest-high-performance-computing