HLRS, alongside its partners in the Gauss Centre for Supercomputing (GCS), is pleased to be participating in ISC High Performance 2024, Europe's largest convention for the high-performance computing (HPC) community. Members of HLRS will be onsite to discuss many of the center's current activities and projects.
Booth Location: K02 Information: https://app.swapcard.com/widget/event/isc-high-performance-2024/exhibitor/RXhoaWJpdG9yXzE1NjY0Mjg=
HLRS will once again appear together with our partners in the Gauss Centre for Supercomputing — the Jülich Supercomputing Centre and Leibniz Supercomputing Centre — in a shared GCS booth. Stop by to learn about our recent activities, and pick up a copy of the HLRS Annual Report or InSiDE Magazine.
This tutorial was developed in the scope of EuroCC 2. As a coordinating center, HLRS supported the organization of the tutorial.
Sunday, May 12 9:00 AM to 1:00 PM Hall Y1 - 2nd floor Information: https://app.swapcard.com/widget/event/isc-high-performance-2024/planning/UGxhbm5pbmdfMTgyNTYyNQ==
Description: Generative Artificial Intelligence (Generative AI) has emerged as a transformative force in creating realistic data, images, and content. By leveraging deep learning (DL) techniques, generative AI models learn patterns and structures from large datasets and autonomously produce novel outputs without explicit programming for each possible outcome. Even though these models are usually pre-trained on massive datasets, fine-tuning for specific tasks or performing inference still requires considerable computational resources, necessitating strategic optimizations to unlock their full potential. This tutorial, powered by EuroCC 2, aims to equip participants with a comprehensive understanding of the computational challenges inherent in generative AI. Moreover, it seeks to present strategies for identifying and mitigating bottlenecks, exclusively leveraging PyTorch’s native features without resorting to additional languages like C++ or CUDA. The tutorial encompasses a diverse array of topics, including hands-on sessions on setting up and optimizing a generic DL pipeline, an exploration of mainstream generative AI models, and optimization strategies for inference specifically tailored to generative AI. Additionally, it investigates advanced subjects such as sparsification and quantization of models with a large number of parameters. By the end of this tutorial, participants will be capable of navigating the challenges of accelerating generative AI using the powerful tools and techniques inherent in PyTorch.
Oleksandr Shcherbakov from HLRS will be one of the speakers of this tutorial.
Sunday, May 12 2:00 PM to 6:00 PM Hall Y9 - 2nd floor Information: https://app.swapcard.com/widget/event/isc-high-performance-2024/planning/UGxhbm5pbmdfMTgyNTYyOA==
Description: The compute continuum is an emerging concept for distributed (and typically centralized) e-infrastructures comprising a set of resources from the edge up to the cloud and HPC. Such infrastructures are needed for various high-impact applications. One example involves collecting observables from IoT devices to improve weather simulations running on HPC systems. Another is in the development of smart cities, where edge-controlled devices are being coupled to compute-intensive machine-learning processes. A key enabling technology is container orchestration, which facilitates deployment of software components and services on both edge devices and cloud infrastructures. The goals of this tutorial are to create a basic understanding of the compute continuum, to improve the knowledge of relevant technologies, and to train their use by deploying and adapting workflows in a compute continuum. The participants will learn about technologies such as containers, the container orchestration solution Kubernetes, an extension towards the edge called KubeEdge, and Prometheus, a software solution for service monitoring. Based on this knowledge, attendees will acquire hands-on experience by implementing a workflow deployed as containers on remote server and local edge nodes.
Booth Location: J30 Information: https://app.swapcard.com/widget/event/isc-high-performance-2024/exhibitor/RXhoaWJpdG9yXzE1NjYyOTc=
Description: As coordinating center of the projects EuroCC 2, CASTIEL 2, and EXCELLERAT P2, and as a project partner in CEEC, HiDALGO2, ChEESE, Inno4scale, HANAMI, and HPC SPECTRA, HLRS will also be present at the EuroHPC Joint Undertaking’s booth, highlighting efforts to build and network HPC expertise across Europe. During the exhibition, several project presentations will be given and videos shown at this booth.
Booth Location: M11 Information: https://app.swapcard.com/widget/event/isc-high-performance-2024/exhibitor/RXhoaWJpdG9yXzE1NjYyNTA
Description: The project bwHPC coordinates support for HPC users across the State of Baden-Württemberg and is implementing related measures and activities for data intensive computing and large-scale scientific data management. The project provides a state-wide HPC infrastructure, federated user support in seven competence centers, and a training program. It is also working to embed current and future data storage systems into the federated HPC infrastructure to unify computing and data services from the user's point of view.
A representative of HLRS will present visualization examples resulting from this project at the bwHPC booth.
On display:
Monday, May 13, 3:00 PM to Wednesday, May 15, 4:00 PM Foyer D-G 2nd Floor Information: https://app.swapcard.com/widget/event/isc-high-performance-2024/planning/UGxhbm5pbmdfMTgyNTA5NQ
Description: In this poster we present the goals and current status of the project CIRCE. The project aims to develop and set up a computational immediate response center for emergencies that can leverage HPC resources. Based on experiences gathered during the COVID-19 pandemic, CIRCE is evaluating the German public sector’s need for HPC and high-performance data analytics (HPDA) resources in emergency situations such as a pandemics, extreme weather events, and disaster management and prevention. We present results obtained so far from the evaluation period along with an overview of the status of two prototypical implementations, which demonstrate CIRCE’s capabilities and evaluate resource requirements in terms of staff and computing time. Based on these evaluations, a concept for the permanent establishment of a computational immediate response center for emergencies at HLRS is developed.
Thursday, May 16 9:00 AM - 1:00 PM Hall Y6 - 2nd floor Information: https://app.swapcard.com/widget/event/isc-high-performance-2024/planning/UGxhbm5pbmdfMTgyMzczOQ==
Description: This workshop is organized in conjunction with ISC24. In a paper session chaired by Francois Mazen, HLRS’s Dennis Grieger will present his paper on interactive in situ visualization, work resulting from the EXCELLERAT P2 project. In it, he explains how the parallel visualization framework Vistle has been extended with a SENSEI-based in situ module. Vistle’s architecture allows an interactive configuration of the visualization pipeline during simulation runtime, also from within immersive virtual environments. Additionally, Vistle handles in-transit data transfer to secondary clusters. First benchmarks on small-scale simulations show promising results in terms of total runtime overhead, as long as the visualization pipeline has enough CPU resources to complete before the next iteration must be analyzed.
Thursday, May 16 9:00 AM - 1:00 PM Hall Y3 - 2nd floor Information: https://app.swapcard.com/widget/event/isc-high-performance-2024/planning/UGxhbm5pbmdfMTgyMzc2OA
Description: Providing a sustainable path for supercomputing is a pressing topic for our community, industry, and governments. Supercomputing has an insatiable appetite for computational cycles, while we face increasing challenges of delivering performance per Watt advances with silicon technology trends. All within the context of climate change, the drive towards net-zero, and economic pressures driven by geo-political challenges. Improving the sustainability of supercomputing provides many opportunities when the end-to-end cycle is considered. From the design of computational circuits and systems; to the power and cooling that is used to operate them, along with the suite of software tools used to administrate, maintain, and raise operational efficiency of HPC systems. All elements of the system must be considered, from compute nodes, and interconnects to IO and storage components of the system. This workshop will bring together users, researchers, hardware and software developers to address opportunities and challenges of sustainability in the supercomputing context.
HLRS research scientist Martin Rose will be one of the speakers at this workshop.
Congress Center Hamburg Congressplatz 1 20355 Hamburg
12. Mai 2024 09:00
16. Mai 2024 13:00
Hamburg
Zurück zur Liste
