HiDALGO is the centre of excellence with the goal of establishing a baseline for HPC, HPDA and AI-oriented computing in the domain of Global Challenges. HiDALGO builds up coupled simulations for highly complex phenomena, focusing on three HiDALGO “pilots”: Simulations of Migration, Social Network Analysis, and Urban Air Pollution. Please visit our main website for more information.
The HiDALGO Training activities are available in this Moodle course repository.
Training videos on the following HiDALGO use cases:
- Urban Air Pollution
- Migration and COVID
- Social Networks
- WP3: Exascale HPC and HPDA System Support
- An overview video
- A hands-on video:
- Time: ca. 45 mins – 1h10 mins.
- Simulation on one aspect of the use-case.
- Participants should apply for remote access.
This course was part of the HiDALGO Online workshop on “Tackling Global challenges with HPC, HPDA and simulations” (7th-9th July 2021). This course has been done by Bernhard Geiger (Know-Center
In this short tutorial, we introduce the concept of a network (or graph) in its most common incarnations. We will discuss graph properties on a micro-level (e.g., node properties such as degree and centrality), meso-level (e.g., concepts such as clique or a community), and macro-level (such as graph diameter and assortativity). All these concepts will be illustrated with practical examples. In the hands-on part of this tutorial, the participants get the chance to construct graph objects from raw data, visualize such graph objects, and compute graph properties using existing libraries. Thus, at the end of the tutorial the participants have obtained:
- a basic understanding of the concept of a graph,
- knowledge of available quantitative measures for network analysis,
- capabilities to visualize graphs,
- capabilities to transform a graph into equivalent representations and vice versa, as well as
- a basic understanding of Python libraries for network analysis.
Prerequisites for this tutorial include basic maths skills and, for the hands-on part, basic programming skills in Python and the ability to execute a provided Jupyter Notebook.
Material of the online course on April 27, 2021:
This training event will start with
an introductory talk to provide a view of high-performance data
analytics (HPDA) from the HiDALGO perspective. The main concepts will be
presented, listing the tools that have been used, together with
information about benchmarks the consortium has done (as a source of
information about their scalability). This introduction also presents
how these tools are being applied in HiDALGO, in order to solve
The following part of the training will focus on HPC and HPDA technologies, applied to use-cases such as Urban Air Pollution (UAP). The UAP application is a software framework for modeling the vehicular traffic emitted air pollution and its dispersion at very high resolution by using geometry inputs (Open Street Map), coupled weather data (ECMWF) and traffic simulation (SUMO), computational fluid dynamics (CFD) tools running on HPC infrastructures (OpenFOAM), and evaluation with HPDA methods.
This HPC/HPDA/UAP-part of the training will introduce the UAP concept, workflows, implementations, application of the CFD-module in HPC environment, deployment to HPC, running, and HPDA for evaluation and model order reduction. Participants will learn the techniques of these parts from a general perspective, namely, HPC workflow modeling (TOSCA in YAML rendering), basics of OpenFOAM for computation of air pollutant dispersion using HPC, and the applied HPDA methods for fast evaluation and model reduction (POD with SVD).
The last part will provide an introduction to the data available at ECMWF and Copernicus, and the APIs for retrieving the data, followed by practical sessions on data exploration and manipulation. After this web-seminar, participants will be able to independently discover weather, climate, and environmental data produced and hosted by ECMWF, and also to retrieve and process these data using Python libraries.
The hands-on part will be carried out using the PSNC (https://www.psnc.pl/) training cluster.
This course is an introduction to the migration pilot:
- Flee and agent-based simulation for forecasting forced migration,
includes a tutorial given at the HiDALGO online workshop 7-9 July 2021
- Designing and prototyping your own simulation with Python3
- Modelling migration on supercomputers
- MUSCLE 3
This seminar is about simplifying user software installation on versatile clusters/testbeds, as well as creating reproducible software environments and benchmarks with Spack and Ansible. It is relevant for any teams involved in software installation and benchmarking.
Sergiy Gogolenko at HLRS, April 14, 2021.
Material from an AI workshop with Huawei
engineers organised by PSNC.
It took place on Thursday, March 4 2021 at 11:00 CET.
• Computing cluster architecture
• Workload managers
• Job management
- Cloudify is the Orchestrator to manage the workflow of applications
- Introduce the basic terminology
- Introduce basic HelloWorld blueprint for running MPI Hello World application in HPC cluster
- CKAN is the data management tool for managing applications' input and output data
- CKAN web options to transfer files using GUI
- CKAN REST API to transfer files using CLI
This in turn leads to a worsened learning experience. It can be helpful if the simulation results are visualized in advance or just in time, so that students can better interact with them.
Recognizing this challenge, we are excited to introduce a cutting-edge solution that revolutionizes the way we interact with CFD results.
Our web-based software, crafted using the latest technologies including Emscripten, OpenGL ES 2.0, SDL2, and IMGUI,
- A software running on the web (WASM), used to visualize CFD results. Only used to visualize data, not to pre-process it (done separately in paraview)
- Written mostly C99-style, compiled as C++
- Compiled with Emscripten
- Rendering done with OpenGL ES 2.0
- Only uses SDL2 and IMGUI as external libraries
- Same look / feel as a desktop app
The HiDALGO2 Urban Building Pilot integrates Dynamic Energy Simulation with indoor air quality models for urban buildings, leveraging advanced algorithms and high-performance computing. Focused on both city-scale and building-scale analyses, it aims to enhance energy efficiency, comfort, and sustainability. This project aligns with global initiatives like the European Green Deal, emphasizing a holistic approach to urban environmental challenges.
This training course will focus on the following points:
- Enhance energy performance, indoor air quality, and human comfort in urban buildings.
- Focus on computational methods for modeling and optimizing urban environments.
Technical Workflow and Challenges:
- Comprehensive workflow integrating mesh generation, parallel mesh adaptation, and multi-fidelity models.
- Challenges in large-scale mesh generation for accurate computational modeling.
- Development of scalable solutions for complex urban building simulations.