<span class="var-sub_title">HPC Meets Real-Time Data: Interactive Supercomputing for Urgent Decision Making</span> SC18 Proceedings

The International Conference for High Performance Computing, Networking, Storage, and Analysis

HPC Meets Real-Time Data: Interactive Supercomputing for Urgent Decision Making


Authors: Nick Brown (University of Edinburgh), Achim Basermann (German Aerospace Center), Vyacheslav Olshevsky (KTH Royal Institute of Technology), Andreas Gerndt (German Aerospace Center), Stefano Markidis (KTH Royal Institute of Technology)

Abstract: Big data and digitalization are creating exciting new opportunities that have the potential to move HPC well beyond traditional computational workloads. Bearing in mind the availability of fast growing social and sensor networks, the combination of HPC simulation with high velocity data and live analytics is likely to be of great importance in the future for solving societal, economic and environmental problems. The idea of this BoF is to bring together those interested in the challenges and opportunities of enhancing HPC with real-time data flows and interactive exploration capacities to support urgent decision making.

Long Description: Through climate change and associated global warming, natural disasters such as the wild fires, hurricanes, and extreme floods, are increasing. Global diseases, e.g., the recent Ebola outbreak, claim scores of victims worldwide every year. Economic loss accrues each year into the trillions of dollars due to the impact of space weather: geomagnetic storms cause disruptions of power grids and communications on the Earth, while solar energetic particle events endanger human lives and satellite operations in space.

With the development and advances in HPC come enhanced opportunities to model and simulate physical, social or economic phenomena, and so to enable better predictions and greater control of such situations. Combining traditional simulations with high velocity sensor data (gathered, for example, from Internet-of-things (IoT) sensors or satellites) enables correlations to be made between simulations and observations such that much more precise and reliable predictions can be generated that feeds into disaster recovery or even prevention.

Our ability to capture data continues to grow exponentially, but often the role that HPC can play in complimenting this and turning data into valuable collateral is overlooked. Combining HPC computational models with real time data can significantly aid in urgent decision making which ultimately saves lives and reduces economic loss. HPC is not just useful from the simulation perspective, but, additionally, the raw computational power provided by supercomputers is of huge value for analysing and visualising vast amounts of raw data in a real-time fashion.

This is not a simple problem to solve and doing this right, in a way that can address the challenges identified requires many technological advances to realise. For instance, given the unpredictable nature of data and its arrival, there are questions around how to take the best advantage of HPC for real-time data fusion, analysis and visualisation as these activities involve highly unpredictable, irregular workloads that must be processed quickly and produce verifiable results. Another major question around is how to organize user interaction such that individuals can reliably interpret their results and make confident decisions in a timely and efficient manner.

This BoF is driven by the VESTEC European project that brings together HPC, data and visualisation experts with application stakeholders in the fields of natural disaster, airborne diseases and space weather domains to solve the challenges identified in this proposal. It is our strong belief that this project should be outward looking, seek to engage with the global HPC community and our technologies should be designed and developed in conjunction with the entire community. Whilst we have three specific problem areas of focus, there are inevitably many more who will benefit from our work and we are keen to engage with other groups and individual researchers tackling similar technological challenges. A BoF at SC is the best opportunity to understand who else globally is involved is tackling these challenges, share with the HPC community what we are doing, and take the first step in building up a community around the fusing HPC with real-time data for interactive supercomputing.





Back to Birds of a Feather Archive Listing