The past few years have seen heavy investments in HPC, which is typically used for solving advanced problems and conducting research activities through computer modelling, simulation and analysis. HPC systems are essential for various applications ranging from defence and security to climate change and personalised medicine, as they allow the implementation of simulation-based predictability with high accuracy. HPC involves thousands of processors working in parallel to analyse billions of pieces of data in real time. It’s crucial to have a new breed of low-power microprocessors to achieve exceptional levels of energy-efficient computer performance. The EU-funded EPI SGA1 project is addressing this issue “to develop the first European HPC system-on-chips (SoCs) and accelerators, with the goal of creating a processor for the Exascale machine based on European technology”, as noted in a news item. “This Exascale supercomputer will be capable of one exaflop of performance – around a million times faster than typical desktop computers – which has the potential to significantly advance AI and scientific research.” FLOPS, a measure of computer speed, refers to the number of floating-point operations per second (FLOPS) that could be performed by a computer. A system delivering one ExaFLOPS would perform 1 billion billion operations per second. This level is expected to be reached by top global players by 2021. Exascale technology will enable far more accurate, detailed, and larger-scale modelling and simulation than those provided by existing systems, as well as new problem-solving approaches such as machine learning and large-scale data analytics.
Launched in 2018, the EPI SGA1 (SGA1 (Specific Grant Agreement 1) OF THE EUROPEAN PROCESSOR INITIATIVE (EPI)) project is the first phase of the European Processor Initiative Framework Partnership Agreement. The EPI is part of a broader strategy implemented by the EU through its EuroHPC Joint Undertaking public-private partnership. “EPI will design and develop the first European HPC System for the HPC and automotive markets through several major streams of operation,” as noted in a news release on the project website. These include a HPC general purpose processor, an accelerator and an automotive platform. The news release states: “Drawing on the expertise of the partners in the consortium, EPI aims to bring a low-power microprocessor to market. It will ensure that the key competence of high-end chip and system design remains in Europe, a critical requirement for many application areas.” Project partners believe that in addition to HPC, EPI’s application areas will cover the autonomous vehicles industry and the data centre and servers market. Quoted in the same news release, Prof. Mateo Valero Cortes, Director at project partner Barcelona Supercomputing Center, says: “Acceleration is crucial to continued performance gains while reducing power consumption in computing. In EPI, the first accelerator will begin from RISC-V technology to deliver two unique vector and artificial intelligence accelerators for HPC and AI, since future supercomputers will be mostly heterogeneous; the second accelerator, based on Kalray’s IP, will lead the path to deterministic automotive computation. Both are offering a European solution to future global converged (HPC and AI) computing needs.” For more information, please see: EPI SGA1 project website