Bernhard C. Geiger and Ali Al-Bashabsheh

On Functions of Markov Random Fields

Hardware in various environments such as High Performance Computing, the Internet of Things and Embedded Systems has become heterogeneous in order to improve computational performance. Customising the hardware for particular application domains as well as the use of accelerators such as GPUs, TPUs, DSPs, FPGAs is attractive as it can lead to performance improvements of up to three orders of magnitude compared to general-purpose processors.

Assisting Decision Makers To Solve Global Challenges With HPC Applications – Covid-19 Modelling

CHALLENGE:​

The current pandemic situation has increased the need of supporting tools to detect, predict and even prevent the virus spread behaviour. Knowing in advance this information will support them to take the appropriate decisions while considering health and care capabilities. In addition, the advance warning of new pandemic waves (or when they may subside) can help health authorities to rescale the capacity for non-urgent care, and ensure the timely arrangement of surge intensive-care capacity.

Success Stories

Each of the success stories below is a summary of a successful work that has been conducted within one of our pilot applications and one or more partners from industry, society and / or science are involved. The summary focuses on the business benefits resulting from the work.

Assisting Decision Makers To Solve Global Challenges With HPC Applications – Migration Issues

CHALLENGE:​

At the same time of the current COVID-19 pandemic, other crises have not stopped like forced migration due to conflicts. In fact, the number of forcibly displaced people is still very high, with over 70 million persons being forced to leave their homes. Save the Children provides support in these countries and needs more accurate estimations on people flows and even destinations to send the appropriate amount of help to the right place.

Robert Elsässer from HiDALGO will give a guest lecture, titled "Spectral methods to compute "eigenvalue histograms" (estimates of eigenvalue frequencies) for graph Laplacians and other LA apps that rely on iterative solvers (e.g in PETCs).", during the Iterative Solvers for Linear Systems course at HLRS on March 8-10, 2021.

HIDALGO researchers presented findings of their studies on „Tackling global challenges with HPC High Performance Computing"

A group of scientists from the EU-funded project HIDALGO collaboratively organised a workshop to discuss pressing issues like the spread of diseases (Covid-19), climate change, air pollution, forced migration, or false and misleading information on twitter and other channels. The speakers showed that HPC, HPDA and Artificial Intelligence could form a pathway to accurately model, simulate, and also to provide forecasts (e.g.

HiDALGO co-organises the "First Joint CoEs Technical Workshop" together with the Centres of Excellence EXCELLERAT and HiDALGO.
This event, which takes place on 27-29 January 2021, is open to HiDALGO, ChEESE and EXCELLERAT partners. Members of other CoEs may also participate.

Several HiDALGO partners present in this event:

Load Balancing session
Robert Elsässer (PLUS): "On Discrete Load Balancing with Diffusion Type Algorithms"

"The HiDALGO project focusses on modelling and simulating the complex processes which arise in connection with major global challenges. The researchers have developed the Flu and Coronavirus Simulator (FACS) with the objective to support decision makers to provide an appropriate response to the current pandemic situation taking into account health and care capabilities...." 

Graph Analyzer Tool

The Graph Analyzer is a validation tool for social graph generators. It compares an empirical graph of a real world social network with a synthetic graph from a generator. It does this by computing and comparing statistical properties of both graphs. For comparison we use distributions of: vertex degree, clustering coefficient, shortest distance between vertices and harmonic centrality.