Home

The management and analysis of modern, large-scale datasets generated by scientific experiments and numerical simulations, can be very challenging due to continuously increasing sizes and complexity. Traditional data mining and analysis methods often rely on computationally complex algorithms which can be very expensive if employed for large-scale datasets. Visual exploration and discovery can then represent invaluable tools, e.g. by providing scientists with prompt and intuitive insights enabling them to identify interesting characteristics and thus define regions of interest within which to apply time-consuming methods. Additionally, they can be a very effective way in discovering and understanding correlations in data patterns, or in identifying unexpected behaviours, thus saving valuable resources, e.g. by terminating promptly on-going numerical simulations producing unreliable results. Visual exploration and discovery tools can also provide effective means for communicating scientific results not only to researchers but also to members of the general public.

Splotch is an open source TRL 6 scientific visualisation software package written in C++, with minimal external dependencies, and suitable for large particle datasets, such as those coming from cosmological N-body simulations. Splotch is implemented with a variety of parallel programming models including hybrid OpenMP and MPI for x86_64 environments such as Xeon and Xeon Phi CPUs, as well as CUDA for GPU-based systems. The lack of dependencies and inclusion of multiple parallel models means that Splotch can be built to exploit almost any type of system, from a stand-alone desktop system with or without a GPU to large heterogeneous HPC architectures. The key strengths of Splotch are support for very large data sets and diverse HPC systems and fast, high-quality volume rendering outputs. Recent developments include moving from a traditional batch software approach to an interactive client–server model. Splotch execution is also supported through appropriate TRL8 cloud services in the emerging European Open Science Cloud (EOSC) ecosystem.

The Splotch development team is composed by:

• Klaus Dolag (Munich Astronomical Observatory)
• Tim Dykes (HPE HPC/AI EMEA Research Lab)
• Claudio Gheller (Institute of Radioastronomy, INAF)
• Martin Reineke (MPI Munich)
• Mel Krokos (University of Portsmouth)
• Marzia Rivi (University College London)

For a quick introduction to Splotch installation and usage, please go to the Getting Started page.