The DiPET project investigates models and techniques that enable distributed stream processing applications to seamlessly span and redistribute across fog and edge computing systems. The goal is to utilize devices dispersed through the network that are geographically closer to users to reduce network latency and to increase the available network bandwidth. However, the network that user devices are connected to is dynamic. For example, mobile devices connect to different base stations as they roam, and fog devices may be intermittently unavailable for computing. In order to maximally leverage the heterogeneous compute and network resources present in these dynamic networks, the DiPET project pursues a bold approach based on transprecise computing. Transprecise computing states that computation need not always be exact and proposes a disciplined trade-off of precision against accuracy, which impacts on computational effort, energy efficiency, memory usage and communication bandwidth and latency. Transprecise computing allows to dynamically adapt the precision of computation depending on the context and available resources. This creates new dimensions to the problem of scheduling distributed stream applications in fog and edge computing environments and will lead to schedules with superior performance, energy efficiency and user experience. The DiPET project will demonstrate the feasibility of this unique approach by developing a transprecise stream processing application framework and transprecision-aware middleware. Use cases in video analytics and network intrusion detection will guide the research and underpin technology demonstrators.

Producing high quality crystals for technological applications is still a challenging task. The manufacturing industry is confronted with complex interdependent physical problems, leading to costly trial-error developments. In parallel scientists, with a broad expertise in numerical simulation and advanced materials science and technology, develop physical models and approaches for numerical modelling. However those developments benefit only to a small number of companies. The core part of the project is to propose a web based marketplace teaching and linking manufacturing industry in the field of crystal growth to materials and process modelling. This will be done through developing a web hub containing an information database to provide the state-of-art in numerical modelling of crystal growth and an open calculation platform. It will allow linking manufacturing industry to modelling community, through high scientific level academic society acting as “translator”. For that, participants in the project, as well as other experts, will offer their portfolio in terms of training/education, numerical simulation, experimental validation, analysis of physico-chemistry of crystal growth processes and physical property measurements. According to our statistics, this could involve, in Europe, up to 150 SMEs and 100 laboratories. The hub will provide an organized set of databases, collecting material properties, comprehensive and convenient bibliography and some examples of typical industrial growth processes simulations. It will also offer integrated software infrastructure, experimental validation procedures for the developed numerical models and experimental data management tools specific to crystal growth processes. This project has the ambition to increase the use of materials simulation by this industrial activity, which provides the basis of many present high technologies. It is likely to increase innovation, reduce costs and decrease time needed for development.

The DiPET project investigates models and techniques that enable distributed stream processing applications to seamlessly span and redistribute across fog and edge computing systems. The goal is to utilize devices dispersed through the network that are geographically closer to users to reduce network latency and to increase the available network bandwidth. However, the network that user devices are connected to is dynamic. For example, mobile devices connect to different base stations as they roam, and fog devices may be intermittently unavailable for computing. In order to maximally leverage the heterogeneous compute and network resources present in these dynamic networks, the DiPET project pursues a bold approach based on transprecise computing. Transprecise computing states that computation need not always be exact and proposes a disciplined trade-off of precision against accuracy, which impacts on computational effort, energy efficiency, memory usage and communication bandwidth and latency. Transprecise computing allows to dynamically adapt the precision of computation depending on the context and available resources. This creates new dimensions to the problem of scheduling distributed stream applications in fog and edge computing environments and will lead to schedules with superior performance, energy efficiency and user experience. The DiPET project will demonstrate the feasibility of this unique approach by developing a transprecise stream processing application framework and transprecision-aware middleware. Use cases in video analytics and network intrusion detection will guide the research and underpin technology demonstrators.