Multiscale phenomena are ubiquitous and they are the key to understanding the complexity of our world. Despite the significant progress achieved through computer simulations over the last decades, we are still limited in our capability to accurately and reliably simulate hierarchies of interacting multiscale physical processes that span a wide range of time and length scales, thus quickly reaching the limits of contemporary high performance computing at the tera- and petascale. Exascale supercomputers promise to lift this limitation, and in this project we will develop multiscale computing algorithms capable of producing high-fidelity scientific results and scalable to exascale computing systems. Our main objective is to develop generic and reusable High Performance Multiscale Computing algorithms that will address the exascale challenges posed by heterogeneous architectures and will enable us to run multiscale applications with extreme data requirements while achieving scalability, robustness, resiliency, and energy efficiency. Our approach is based on generic multiscale computing patterns that allow us to implement customized algorithms to optimise load balancing, data handling, fault tolerance and energy consumption under generic exascale application scenarios. We will realise an experimental execution environment on our pan-European facility, which will be used to measure performance characteristics and develop models that can provide reliable performance predictions for emerging and future exascale architectures. The viability of our approach will be demonstrated by implementing nine grand challenge applications which are exascale-ready and pave the road to unprecedented scientific discoveries. Our ambition is to establish new standards for multiscale computing at exascale, and provision a robust and reliable software technology stack that empowers multiscale modellers to transform computer simulations into predictive science.

The overall objective of the Next Generation I/O Project (NEXTGenIO) is to design and prototype a new, scalable, high-performance, energy efficient computing platform designed to address the challenge of delivering scalable I/O performance to applications at the Exascale. It will achieve this using highly innovative, non-volatile, dual in-line memory modules (NV-DIMMs). These hardware and systemware developments will be coupled to a co-design approach driven by the needs of some of today’s most demanding HPC applications. By meeting this overall objective, NEXTGenIO will solve a key part of the Exascale challenge and enable HPC and Big Data applications to overcome the limitations of today’s HPC I/O subsystems. Today most high-end HPC systems employ data storage separate from the main system and the I/O subsystem often struggles to deal with the degree of parallelism present. As we move into the domain of extreme parallelism at the Exascale we need to address I/O if such systems are to deliver appropriate performance and efficiency for their application user communities. The NEXTGenIO project will explore the use of NV-DIMMs and associated systemware developments through a co-design process with three ‘end-user’ partners: a high-end academic HPC service provider, a numerical weather forecasting service provider and a commercial on-demand HPC service provider. These partners will develop a set of I/O workload simulators to allow quantitative improvements in I/O performance to be directly measured on the new system in a variety of research configurations. Systemware software developed in the project will include performance analysis tools, improved job schedulers that take into account data locality and energy efficiency, optimised programming models, and APIs and drivers for optimal use of the new I/O hierarchy. The project will deliver immediately exploitable hardware and software results and show how to deliver high performance I/O at the Exascale.

ESiWACE will substantially improve efficiency and productivity of numerical weather and climate simulation on high-performance computing platforms by supporting the end-to-end workflow of global Earth system modelling in HPC environment. This will be obtained by improving and supporting (1) scalability of models, tools and data management on state-of-the-art supercomputer systems (2) Usability of models and tools throughout the European HPC eco-system, and (3) the Exploitability of the huge amount of resulting data. We will develop solutions for cross-cutting HPC challenges particular to the weather and climate domain. This will range from the development of specific software products to the deployment of user facing services for both, computing and storage. ESiWACE leverages two established European networks, namely (1) the European Network for Earth System modelling, representing the European climate modelling community and (2) the world leading European Centre for Medium-Range Weather Forecasts. The governance structure that defines the services to be provided will be driven by the European weather and climate science community. Weather and climate computing have always been one of the key drivers for HPC development, with domain specific scientific and technical requirements that stretch the capability and capacity of existing software and hardware to the limits. By developing solutions for Europe and at European scale, ESiWACE will directly impact on the competitiveness of the European HPC industry by engendering new products, providing opportunities for exploitation beyond the project itself, and by enhancing the skills base of staff in both industry and academia. ESiWACE will be at once thematic, as it focuses on the HPC application domain of climate and weather modeling, transversal, as it covers several aspects of computational science, and challenge-driven, as climate and weather predictability represents a major societal issue.