Interview with Sai Narasimhamurthy
After Exascale: SPE opens the discussion on Europe’s next computing choices
AI factories, Giga-Factories, frontier models, RISC-V processors, quantum accelerators, energy-hungry data centres: the post-Exascale debate is no longer limited to faster machines. It now touches infrastructure, software, data, and Europe’s place in a rapidly shifting global landscape.
These are some of the issues at the heart of the Strategy for Post-Exascale (SPE) project. In this interview, Sai Narasimhamurthy, SPE Work Package 1 leader, Technology Roadmapping leader at ET4HPC and Senior Technologist at ParTec shares his thoughts on how SPE is going to help build an European roadmap and give policy recommendations for the future of advanced computing.
What is changing in advanced computing, and why does Europe need to look beyond Exascale now?
The computing landscape is being reshaped by the convergence of HPC, AI and Quantum Computing and Simulation. AI is the most visible disruption today, but it does not stand alone. It relies on large-scale infrastructures, accelerators, software stacks, data pipelines, storage systems and energy capacity.
Europe faces a practical question: where do we put our effort? The United States is investing huge amounts into AI infrastructures, with a strong role for private capital. China is also moving fast. Europe cannot simply copy that model or match it euro for euro.
The choice is therefore not only technological. Europe needs to identify where it can build excellence, where it must reduce dependency, and where collaboration with global partners makes more sense than trying to act alone.
« Europe needs to identify where it can build excellence. »
Why is sovereignty such a central issue in this context?
Sovereignty does not mean isolation. It means that Europe should not be overly dependent on technologies, infrastructures or models that it cannot control.
One obvious example is AI hardware. European data centres still rely heavily on processors, accelerators and GPUs from non-European companies. These technologies are essential, and collaboration with established suppliers remains valuable. But the dependency is real. This is where initiatives such as DARE, developments around RISC-V, and European companies such as SiPearl become part of the wider picture. They do not solve the issue overnight. They show that Europe is starting to build parts of its own hardware ecosystem.
Europe can progress in stages. Software is an area where we can reduce dependencies more quickly, especially by supporting open-source approaches and building on our strong expertise. Europe has significant strengths in system software, research and collaborative projects. These will be essential for the future of HPC and AI infrastructures.
Another key aspect is data. With the rise of AI, storage, I/O and data management are becoming critical. Data is the fuel of the AI revolution. If Europe wants to build competitive infrastructures, we must invest in better data management frameworks and storage technologies.
« SPE is about preparing a roadmap for the next Multiannual Financial Framework. »
How does the SPE project aim to address these challenges?
SPE will go further than the Strategic Research Agenda (SRA) developed by ETP4HPC by looking at a longer time horizon. While the SRA typically focuses on the next few years, SPE is about preparing a roadmap for the next Multiannual Financial Framework, from 2028 to 2034.
The roadmap is intended to inform European funding priorities, policy decisions and strategic investments over that period, helping align research, industry and infrastructure development with long-term objectives. It will not be created in a single step. We will start with an initial vision, then continuously refine it through stakeholder input, workshops and discussions with policymakers. This iterative approach allows us to integrate feedback and remain open to new and disruptive ideas.
Will SPE also address how advanced computing is funded in Europe?
Beyond defining the technological framework for post-Exascale computing, SPE will also develop policy recommendations and funding priorities. Today, European research funding often combines support from the European Commission or EuroHPC with matching contributions from Member States. While this model enables large-scale collaboration, it is frequently seen as complex and bureaucratic.
As part of its roadmap work, SPE will examine how these mechanisms could be streamlined to make funding more accessible and effective. This includes identifying ways to reduce administrative burden and improve coordination between European and national programmes. The project will therefore not only outline technological directions, but also propose funding recommendations to better support research, innovation and deployment in advanced computing.
« SPE will examine how to make funding more accessible and effective. »
What would make SPE successful?
Two signs will matter. The first will take time to assess: whether SPE’s roadmap, policy recommendations and policy papers influence future European priorities and funding programmes, especially in the next Multiannual Financial Framework.
The second is more immediate: whether SPE succeeds in bringing together the many existing communities across Europe that today often work in parallel or in an ad hoc way. One of the project’s key objectives is precisely to build a strong European community of stakeholders. In that sense, there is already a form of success built into the process: for the first time, these different communities are coming together to collectively develop a post-Exascale roadmap. This creates cohesion and establishes stronger links between actors, which can support Europe’s move towards greater sovereignty. Seeing these connections take shape and strengthen over the course of the project would already be a clear and immediate indicator of progress.
Fill in the form
Join Us to Help Shape the Future of Computing
SPE project
This project has received funding from the European Union’s Horizon Europe research and innovation program under grant agreement N°101296624. This website reflects only the author's view and the Commission is not responsible for any use that may be made of the information it contains.