The European High Performance Computing Joint Venture (EuroHPC JU) has announced a major new funding round to reshape the way High Performance Computing (HPC) systems are benchmarked and evaluated across Europe.
This initiative, backed by a budget of €4 million under the Horizon Europe program, aims to establish a robust, transparent and future-proof benchmarking methodology that can keep pace with rapidly evolving computing technologies.
Accepting submissions until 17:00 CET on March 24, 2026, this call invites researchers, industry players, and technology organizations to contribute to a new generation of benchmarking frameworks for both traditional and emerging hybrid computing environments.
The importance of high performance computing
High-performance computing refers to the use of powerful computing systems that can process large amounts of data and perform complex calculations extremely quickly.
HPC systems are essential for tasks beyond the capabilities of traditional computers, such as climate modeling, drug discovery, aerospace simulation, artificial intelligence training, and advanced materials research.
The potential for HPC continues to grow as Europe invests heavily in exascale computing and AI-driven workloads.
Well-designed benchmarks can fully realize this potential by helping users match the right system to the right application, while fostering innovation in energy-efficient and sustainable computing architectures.
Building a unified benchmark standard for HPC systems
Central to this effort is the ambition to create a unified, modular, and extensible benchmarking framework for HPC systems.
EuroHPC JU aims to replace fragmented approaches with a well-documented, continuously updated set of benchmarks that can be widely adopted across Europe.
Benchmarks play an important role in high performance computing, allowing fair and reproducible comparisons of hardware and software platforms.
A common framework allows users to verify performance claims, compare systems objectively, and make informed decisions based on real-world workloads rather than individual metrics.
For system operators, this effort also supports optimization of throughput and energy efficiency. These two priorities are becoming increasingly important as HPC systems scale toward exascale performance.
Focus Area 1: Benchmarking Exascale and HPC-AI Systems
The first topic of this call, European HPC-centric benchmarking frameworks, targets exascale computing and HPC-AI convergence.
The proposal is expected to define and standardize benchmarks that will enable reproducible, reproducible, and replicable performance evaluations across different HPC systems.
Key goals include the development of fine-grained performance indicators, particularly related to energy efficiency, and the standardization of benchmark inputs and outputs.
Applicants must also design a structured workflow that captures the entire benchmark lifecycle, from deployment to data collection and analysis.
Focus Area 2: Benchmarking Hybrid Quantum-Classical Computing
The second topic shifts the focus to the future and covers benchmarks for hybrid quantum-classical computing.
As quantum technology matures, there is growing interest in integrating quantum processors with traditional HPC systems.
Proposals in this area should define hardware-independent benchmarks that measure throughput, latency, energy consumption, and the depth of integration between HPC and quantum computing.
Applicants are also expected to outline a strategic roadmap that anticipates technological advances and supports diverse qubit modalities.
Strengthening Europe’s HPC ecosystem
By investing in advanced benchmarking frameworks, EuroHPC JU is strengthening Europe’s leadership in HPC systems while ensuring transparency, comparability and sustainability.
This conference represents an important opportunity to shape how performance is measured in hybrid and quantum-enabled HPC environments, not only today but in the future.
Source link
