Categories
Pages
-

IT Center Events

Schlagwort: ‘HPC’

VI-HPS Porting and Tuning Workshop 2024

December 28th, 2023 | by
Date and Time: 26th Feb 2024 8:00 am – 1st Mar 2024 6:00 pm

Location: IT Center RWTH Aachen & ZIH, TU Dresden

Klick here for Registration and further information.

 


 

Description

This workshop is organized by VI-HPS and the NHR centers NHR4CES@RWTH (IT Center, RWTH Aachen University) and NHR@TUD (ZIH, TU Dresden) and focuses on the new computing platforms “CLAIX-2023” and “Barnard” at RWTH Aachen and TU Dresden, respectively.

The workshop will be co-scheduled with the workshop of the same name at NHR@TUD in Dresden. Presentations will be streamed to the corresponding other location depending on the location of the speaker.

Tool experts for the hands-on tuning part will be on-site at both locations. Note that no hands-on tuning assistance can be given to online-only participants. For on-site hands-on participants you will have to bring a self-written application code written in C/C++/Fortran/Python that is parallelized with common HPC programming Paradigms, such as OpenMP, MPI, and/or C++ and POSIX threads.

On completion participants should be familiar with the module system and SLURM configuration and monitoring systems on CLAIX-2023 and Bernard (depending on selected site), common performance analysis and diagnosis techniques and how they can be employed in practice (on a range of HPC systems). Those who prepared their own application test cases will have been coached in the tuning of their measurement and analysis, and provided optimization suggestions.

Target Audience:

Monday (Porting day): Any current and future user who is interested in building and/or running software on the prospective HPC system of RWTH Aachen University, CLAIX-2023.

Tuesday–Friday (Tuning days): Any developer of C/C++/Fortran HPC software using OpenMP, Pthreads, C++ threads, and/or MPI, who is interested in identifying performance bottlenecks within the HPC application.

Gained skills:

Monday (Porting day):

– Basic knowledge about the Hardware of CLAIX-2023
– Using the software stack and module system of CLAIX-2023
– Best practices for building software on CLAIX-2023
– Writing SLRUM job submission scripts for CLAIX-2023
– Using the job monitoring system of CLAIX-2023 to verify expected performance parameters of an application run

Tuesday–Friday (Tuning days):

– Identifying performance problems of your application on the node-level using MAQAO and LIKWID
– Gaining an runtime summary of an application using Score-P and TAU
– Using advance performance explorers to navigate the measurement results
– Tuning the Score-P measurement configurations for your application
– Obtaining event trace information for your application using Score-P
– Using Scalasca to automatically search for inefficient application behavior
– Using Vampir for in-depth understanding of process and thread interaction
– Using lo2s to understand execution behavior of an application and OS influence on it

Prerequisites:

Monday (Porting day):

– Knowledge on how to access the current CLAIX-2018 system
– Basic knowledge about the SLURM job scheduling system (join Introduction to HPC for those basic skills)

Tuesday (Tuning days):

– An HPC code written in C/C++/Fortran using OpenMP/Pthreads/C++-threads and/or MPI
– Active developer with in-depth understanding of the provided code
– General understanding of building software on an HPC platform

Note that registration for this workshop for online or on-site participation at either location is handled via the registration page at TU Dresden. It is possible to register for individual days.

Please contact us for any further questions.


 

Agenda

The workshop has two parts with individual focus: The first part focuses on porting your application to the respective local platform at RWTH Aachen and TU Dresden. It will give an overview of the Sapphire Rapids architecture and node configuration, the local environments and monitoring solutions, and how to use the systems effectively. Presentations will take place on the first day, assistance will be given on-site and online on Monday, and online-only for the remainder of the week.

The second part (starting on Tuesday) focuses on tuning your application on the systems with the use of several tools of the VI-HPS programming tools suite. It will give an overview of the selected tools, explain their functionality, and how to use them effectively. Furthermore, it will offer hands-on experience and expert assistance using the tools on participants’ own codes.

Introduction to the HPC JupyterHub at the RWTH

November 13th, 2023 | by

Abstract

The HPC JupyterHub service at the RWTH allows all eligible users of the RWTH Compute Cluster to utilize the existing compute hardware interactively with the use of Jupyter Notebooks. This HPC JupyterHub provides customization of Profiles with a variety of programming kernels, software packages, and hardware definitions. This workshop will showcase the HPC JupyterHub service.

Target Audience

New HPC JupyterHub users: HPC, Simulation Software and Machine Learning users.

Requirements

HPC and VPN account at the RWTH for the interactive demo.

Registration

Registration

NHR4CES: CFD Training Series: Introduction to Kernel-based approximation methods with applications to fluid dynamics

September 28th, 2023 | by

Name: CFD Training Series: Introduction to Kernel-based approximation methods with applications to fluid dynamics

Date: November 23, 2023 

Time: 1.00 pm – 5.00 pm

Format: hybrid

More Information

 


 

Description

Short abstract:

When data is provided in an unstructured format or is high-dimensional, classical interpolation or approximation schemes as Finite-Element Methods (FEM) struggle to be accurate and efficient. An alternative is provided by kernel-based approaches in the Reproducing Kernel Hilbert space (RKHS) framework. Common applications range from support vector machines in context of machine learning to reconstruction of image data. In this course, we will introduce the framework for kernel-based approximation schemes and discuss how one can implement them efficiently. At the end we will look at possible applications of kernel-based methods in context of fluid dynamics and particle methods.

Click here to register online

Click here to register on site

NHR4CES: Data-Driven Healthcare: Leveraging Visualization and Data Management for Digital Patient Studies

September 28th, 2023 | by

Name: Data-Driven Healthcare: Leveraging Visualization and Data Management for Digital Patient Studies

Date: November 21, 2023 & November 22, 2023 

Time: 1.00 pm – 5.00 pm

Format: Online

More Information

 


 

Description

Short abstract:

In today’s rapidly advancing medical landscape, data has become an invaluable asset, empowering healthcare professionals and researchers to make informed decisions, derive insights, and develop innovative solutions. However, coping with the sheer volume and complexity of data while ensuring security and privacy aspects is a daunting task. To harness the full potential of this data-rich environment, effective visualization techniques and robust data management strategies have emerged as indispensable tools.

This NHR4CES workshop seeks to shed light on how professionals and researchers can effectively leverage visualization and data management to tackle current and emerging challenges, supporting medical experts in their daily work, research, and improving communication on and with patients. Researchers and users from different domains are invited to exchange and discuss ideas alongside a program filled with presentations by professionals of different fields as well as panel discussion.

 

Click here to register

NHR4CES: CFD Training Series: Introduction to Discontinuous Galerkin Methods for Flow Problems

September 28th, 2023 | by

Name: CFD Training Series: Introduction to Discontinuous Galerkin Methods for Flow Problems

Date: November 16, 2023 

Time: 1.00 pm – 5.00 pm

Format: Online

More Information

 


 

Description

Short abstract:

In this course, we cover the main building blocks to solve fluid flow problems using the Discontinuous Galerkin (DG) method. The course consists of a combination of presentations and hands-on exercises in which a simple DG flow solver is implemented and run on some test cases within our open-source code framework BoSSS.

 

Click here to register

NHR4CES: Parallelization in OpenFOAM for HPC Deployment

September 28th, 2023 | by

Name: Parallelization in OpenFOAM for HPC Deployment

Date: November 06, 2023 & November 16, 2023

Time: 10am – 4pm

Format: Online Flipped Classroom focusing on hands-on training

More Information

 


 

Description

Short abstract:

OpenFOAM is an open source, mature and established C++ library for computational continuum mechanics (CCM) including Computational Fluid Dynamics (CFD). For leveraging its full potential, it is crucial to efficiently use the high-performance computing (HPC) resources on modern distributed-memory parallel computer architectures. This must be based on a sound understanding of parallelization in OpenFOAM and HPC techniques available.

The training will be concerned with introducing the participants to the different concepts of parallelization, along with code examples for illustration. Moreover, we will provide hands-on exercises to further deepen and solidify the transferred knowledge. The participants will further gain an overview over the distinct techniques and dedicated tools involved to run a massively parallel computation using OpenFOAM, as well as over ongoing HPC-related activities in research and development.

 

Click here to register

NHR4CES: CFD Training Series: Introduction to Turbulence Modeling and Numerical Implementation

September 28th, 2023 | by

Name: CFD Training Series: Introduction to Turbulence Modeling and Numerical Implementation

Date: November 02, 2023 

Time: 1.00 pm – 5.00 pm

Format: online

More Information

 


 

Description

Short abstract:

In this course, the introduction to the structural properties of various turbulence modeling concepts (RANS, LES, and Hybrid RANS/LES) including associated equations will be given. In addition to the presentation, the corresponding computational setup including pre-processing, simulation implementation, and post-processing for some illustrative flow configurations will be provided based on the open-source CFD software OpenFOAM.

 

Click here to register

NHR4CES: CFD Training Series: Efficient HPC implementation for Lagrangian particle tracking

August 30th, 2023 | by

Name: CFD Training Series: Efficient HPC implementation for Lagrangian particle tracking

Date: October 26, 2023 

Time: 1.00 pm – 5.00 pm

Format: hybrid

More Information

 


 

Description

Short abstract:

In this course, the basics of Lagrangian point particle methods for the application on HPC systems are covered. The course consists of an introduction to the applied method, followed by a hands-on exercise using the in-house simulation framework m-AIA. The topics covered are spherical and non-spherical particles and the efficient implementation of point particle methods for the use in HPC. The course will be held in person at the Chair of Fluid Mechanics and Institute of Aerodynamics at RWTH Aachen University in Aachen. The presentations will also be streamed online.

 

Click here to register

 

NHR4CES: Introduction to Machine Learning and Deep Learning

August 30th, 2023 | by

Name: Introduction to Machine Learning and Deep Learning

Date: November 09, 2023 & November 10, 2023

Time: 9 a.m to 1 p.m.

Format: Online

More Information

 


 

Description

Short abstract:

Not only in economics Machine- and Deep-Learning (ML/DL) are inherently used to solve highly complex problems in a data-driven way, but also the scientific community has many use-cases in which ML/DL are useful, e.g. to discover hidden patterns or replace computationally heavy simulations with data-driven  approaches. The participants will learn how to design ML-models by themselves and will learn about possible pitfalls when applying ML in the real  world.

 

Click here to register

 

NHR4CES: Process Mining and Scientific Workflows running on the HPC cluster

August 30th, 2023 | by

Name: Process Mining and Scientific Workflows running on the HPC cluster

Date: December 11, 2023

Time: 9 a.m to 1 p.m.

Format: Online

More Information

 


 

Description

Short abstract: The goal of Process Mining is to turn event data into insights and actions. On the other side, there exist scientific workflows running on HPC clusters. But, why do we need process mining to analyze scientific workflows running on HPC clusters? For two reasons, documentation of scientific workflows and detection of bottlenecks that slow down the execution of scientific workflows. That is why we are doing to implement a cockpit to monitor HPC processes with Process Mining techniques. Another perspective is supporting Process Mining workflows for scientific experiments to facilitate the use and also improve the performance of Process Mining techniques.

 

Click here to register