Last week, members of the EU’s CoE RAISE project met at CERN for their “All Hands” meeting. This innovative project is developing artificial-intelligence (AI) approaches for next-generation “exascale” supercomputers, for use across both science and industry. Use cases explored through the project include the optimisation of wind-farm layouts, design of efficient aircraft, improved sound engineering, seismic imaging with remote sensing, and more.

CoE RAISE, The European Center of Excellence in Exascale Computing “Research on AI- and Simulation-Based Engineering at Exascale”, is funded under the EU’s Horizon 2020 research and innovation programme. The project launched in 2021 and runs for three years. Discussions to plan for a potential follow-on project were also held during last week’s meeting.

The four-day meeting, which took place in CERN’s Council Chamber, was attended by 54 project members. The participants discussed progress made in their work to develop AI technologies for complex applications in Europe running on future “exascale” high-performance computing (HPC) systems. Exascale refers to the next generation of high-performance computers that can carry out over 1018  floating point operations per second (FLOPS). Today, only the Frontier supercomputer at Oak Ridge National Laboratory in the United States has reached this level. However, with more exascale HPC systems just over the horizon, it is important to ensure that AI approaches used in science and industry are ready to capitalise fully on the enormous potential. In June, the European High Performance Computing Joint Undertaking (EuroHPC JU) announced that Forschungszentrum Jülich GmbH in Germany has been selected to host and operate Europe’s first exascale supercomputer, which is set to come online next year and will be known as the Joint Undertaking Pioneer for Innovative and Transformative Exascale Research (JUPITER) system.

CoE RAISE is developing innovative AI methods on heterogeneous HPC architectures involving multiple kinds of processors. Such architectures can offer higher performance and energy efficiency, but code must be adapted to use the different types of processors efficiently. The AI methods being developed are focused around nine key use cases and designed to scale well for running on exascale HPC systems.

CoE RAISE supports technology transfer to industry, particularly small- and medium-sized enterprises, as well as running education and training initiatives. On top of this, CoE RAISE also provides consulting and connects to other European initiatives to maximise synergies, exploit opportunities for co-design and share knowledge. All aspects of the project’s work were discussed over the four days at CERN.

Member organisations in the CoE RAISE project include the Barcelona Supercomputing Center, Forschungszentrum Jülich GmbH, and the Centre Européen de Recherche et de Formation Avancée en Calcul Scientifique (CERFACS). Several leading European Universities are members, as well the companies Bull SAS, a major French IT brand belonging to Atos; ParTec AG, which develops modular supercomputing architectures; and SAFRAN Helicopter Engines.

CERN is also a partner and brings one of the use cases to the project. This work focuses on the improvement of methods for reconstructing particle-collision events at the upgraded High-Luminosity Large Hadron Collider (HL-LHC), which is set to come online in 2029. The HL-LHC will see more particle collisions than ever taking place, producing exabytes of data each year, resulting in unprecedented computing challenges. To reconstruct particle collision events today (with datasets in the order of terabytes or petabytes), hundreds of different algorithms run concurrently: some are traditional algorithms optimised for particular hardware configurations, while others already include AI-driven methods, such as deep neural networks (DNNs). The members of the project team at CERN are working to increase the modularity of systems and ensure code is optimised to fully exploit heterogeneous architectures, as well as increasing the use of machine learning and other AI methods for reconstruction of collisions and classification of particles.

“Supercomputers are reaching the exascale and enabling the delivery of an unprecedented scale of processing resources for HPC and AI workflows,” says Maria Girone, CERN openlab CTO, who leads CERN’s contribution to the project. “The research performed in CoE RAISE will drive the co-design of HPC computing resources for future AI and HPC applications for both science and industry. This meeting enabled us to exchange and develop ideas and to bring new perspectives. It also gave researchers from other domains a unique insight into the environment and challenges facing CERN, enabling cross-fertilisation and understanding.”

“The participants in the meeting – some of them having seen each other for the first time in the project due to COVID-19 restrictions – made new friends and initiated more intensified collaborations on emerging topics, such as the integration of quantum computing in AI and HPC workflows,” says Andreas Lintermann of Forschungszentrum Jülich GmbH, who coordinates the project. “Members went home with new strategies in mind for fostering their implementations and ensuring sustainability, with the goal of creating impact in the final year of the project and beyond.”

— Andrew Purcell

2022 saw the implementation of a new strategy and new structure for the CERN IT department. Under these, a new “Innovation” section has been created, bringing together CERN openlab, CERN Quantum Technology Initiative (CERN QTI) and several IT-focused EU projects and scouting activities, including collaborations with other research organisations. This recognises the vital need to innovate new computing technologies to support the ambitious upgrade programme for the Large Hadron Collider (LHC): when the High-Luminosity LHC (HL-LHC) comes online in 2029, the total computing capacity required by the experiments is expected to be ten times greater than today (read more in the announcement marking the start of building work for the new CERN data centre in Prévessin).

Technical and outreach events kick off a year of important progress

In Q1, CERN openlab held its annual technical workshop. Over 200 people attended this event, which discussed work carried out through 32 joint R&D projects spread across CERN.

In Q2, CERN QTI supported the second ever World Quantum Day celebration, even organising a special scientific symposium at CERN. Talks outlined the early days of quantum science at CERN and what that pioneering effort means for modern research.

The Quantumacy project also came to a successful close at this time. Members of the consortium met in Paris, France, to mark the conclusion of this project, which had been investigating privacy-preserving forms of quantum communication.

On top of this, a new project called SYCLOPS was approved under the new EU Framework Programme, Horizon Europe. This project will work to enable better solutions for AI and data-mining for extremely large and diverse data by democratising AI acceleration using open standards, and enabling a healthy, competitive, innovation-driven ecosystem for Europe and beyond.

Summer students return, as new quantum and AI collaborations forged

Q3 saw two important European research institutes – INFN (Istituto Nazionale di Fisica Nucleare) and IIT (Italian Institute of Technology) – join CERN’s hub in the IBM Quantum Network. Both institutes are now working closely with CERN to help investigate the full potential of nascent quantum-computing technologies, sharing access to IBM’s fleet of quantum computers via the cloud.

A new EU project called InterTwin began work. It will build the prototype for a universal “digital twin” engine, a software programme that is able to virtually replicate any physical device, product or entity thanks to machine learning and software analytics. The project involves a consortium of 28 participants from 12 countries, bringing together modelling and simulation experts in high-energy physics, radio astronomy, astroparticle physics, climate research and environmental monitoring.

The CERN openlab Summer Student Programme returned as a full, in-person event for the first time since the start of the COVID-19 pandemic. Over nine weeks, the students attended specialist lectures and worked on projects involving some of the latest hardware and software technologies. 32 students, representing 19 nationalities, were part of this year’s programme.

This year’s summer students were also joined by the winners of the QHACK quantum machine-learning hackathon and the oneAPI Great Cross-Architecture Challenge, organised by Intel with support from CERN and Argonne National Laboratory in the US.

On top of this, CERN openlab launched its new website, too.

Quantum quest makes quite the quake in quarter “quatre”

The final part of the year was particularly busy for all in the CERN IT department’s Innovation section. The quarter started with the announcement of the Open Quantum Institute. The institute, which was announced in a speech by CERN’s Director-General, will work to ensure that emerging quantum technologies are put to use to tackle key societal challenges. The proposal for this institute was made through GESDA, the Geneva Science and Diplomacy Anticipator Foundation, in collaboration with leading research institutes and technology companies. Other founding supporters of the Open Quantum Institute include the University of Geneva, the Swiss Federal Institutes of Technology in Zurich (ETH) and Lausanne (EPFL), Microsoft and IBM.

CERN QTI organised the International Conference on Quantum Technologies for High-Energy Physics (QT4HEP). This first-of-its-kind event brought together members of the quantum technologies community from research and industry. They discussed recent developments in the field and worked to identify activities within particle physics – and other sciences – that can most benefit from the application of quantum technologies. 224 people attended in person, with more following online. Recordings from the event are available here.

Other news from the final quarter of this pivotal year includes the award of a prestigious Italian prize to CERN openlab’s CTO, Maria Girone, and the transfer of technical coordination of the BioDynaMo project from CERN to the University of Cyprus, as part of the project consortium’s goal to seek new applications for its powerful platform for agent-based modelling. Applications for the 2023 CERN openlab Summer Student Programme were also opened in Q4.

Throughout 2022, CERN made important contributions to an EU project called RAISE, The European Center of Excellence in Exascale Computing "Research on AI- and Simulation-Based Engineering at Exascale".  Work is now underway to help prepare the project’s next “all-hands” meeting, which will be held at CERN on 17-20 January.

Strong foundations for important innovation in 2023

2022 was an important year for all projects and initiatives under the umbrella of the new IT Innovation section. The IT department’s new structure and strategy is already helping to ensure that innovation in computing technologies is well targeted and can have maximum impact across the laboratory and CERN’s wider research community. Priorities in 2023 include:

1. Further work on the integration of cloud and high-performance computing with heterogenous, accelerated computing architectures
2. The design and deployment of state-of-the-art technologies to support AI applications for physics
3. Scaling up our quantum algorithms to new quantum devices, promising noise-free operations for the first time.

“2022 was a critical year of transformation for us, during which we have both rationalised our activities around a few core themes and laid out the foundations for longer-term activities for the HL-LHC and beyond,” says Alberto Di Meglio, the head of IT Innovation. “In 2023, we will focus on co-developing and testing with our user communities the solutions they need. We will also strengthen the collaboration on the future of computing, storage and algorithms for science — not only within particle physics, but also with other communities like climate and environment, Earth observation, and medical applications.” 

— Andrew Purcell