Conference Agenda

Brief Description and Outline:

This workshop introduces the HAICORE HPC Cluster at HZDR, a high-performance computing resource available to Helmholtz AI projects. This session focuses on how researchers can effectively utilize our Open OnDemand web portal to run AI and data-intensive workloads and our comprehensive web application suite for Machine Learning workflows.

We highlight how HAICORE integrates established HPC technologies with modern web services and AI oriented tooling, to enable reproducible and scalable research across Helmholtz centers.

Workshop Outline (2 hours)

- Introduction to HAICORE at HZDR (10 min)
Overview of HAICORE resource allocation portal (ColdFront)
Allocation process, account provisioning, and available resources.

- Getting Started: HAICORE Web Portal (20 min)
Accessing HPC resources through Web-portal (Open OnDemand).
Overview of available application suite for Coding, MLOps, Data Transfer etc.
Introduction to HAICORE Cluster Software stack.

- Running Machine Learning Workloads (40 min)
Training Scientific Deep Learning models.
Using MLOps tools for tracking and reproducibility.

- GenAI Workflows on HAICORE (40 min)
Deploying Language Models on the Cluster.
Using MLFlow for Generative AI workflows.

- Q&A and Discussion (10 min)

Goals:

Workshop Goals

• Introduce Helmholtz AI researchers to brand new HAICORE HPC resources at HZDR.
• Demonstrate the utility of HZDR HPC-labs web portal and its comprehensive application suite.
• Encourage reproducible and scalable AI workflows by leveraging best MLOps practices.

Importance for HAICON26:

As Helmholtz AI initiatives increasingly require scalable GPU resources, reproducible workflows, and support for emerging areas such as Generative AI, there is a strong need to make these capa-bilities accessible and practical for researchers.

This workshop directly addresses that need by guiding participants through the full lifecycle of working on HAICORE —from allocation via ColdFront, to interactive access through Open On-Demand, to executing machine learning and GenAI workflows with MLOps best practices. The workshop supports the broader HAICON 2026 mission of enabling scalable and reproducible AI research across Helmholtz centers.

Presenters Experience:

Varun Sudharshanam:

Scientific Staff at HPC‑Labs, HZDR, focusing on HPC infrastructure and machine‑learning workflows on HPC clusters.

Prior Teaching Experience:

• Conducted multiple High Performance Computing Workshops at HZDR and UFZ.
• Instructor for Multi‑GPU Workloads at the HPC‑Gateway Fall School 2025, HZDR.
• HIDA Instructor for Software Carpentries.

Dr. Kushal Ramakrishna:

Research Scientist at HPC‑Labs, HZDR, advancing ab initio electronic‑structure simulations and

multiscale materials modelling through data‑driven and machine‑learning approaches for materials science.

Prior Teaching Experience:

• Instructor for the lab component of the “Highly Parallel Programming of GPUs” course at the Faculty of Computer Science, TU Dresden.
• Conducted multiple High Performance Computing Workshops at HZDR and UFZ.

Target Audience:

This workshop is aimed at researchers, PhD students, postdocs, and research software engineers

from Helmholtz centers who:

• Develop or apply machine learning / AI methods in scientific research
• Require scalable GPU resources for training or deploying models
• Are interested in reproducible MLOps or LLM-Ops workflows
• Want to transition from local experimentation to production-grade HPC environments.

Participants are expected to have:

• Basic familiarity with Python and machine learning concepts
• Some experience working in Linux environments (helpful but not mandatory)
• No prior experience with HAICORE or HPC systems is required

The session is designed to be accessible to both AI practitioners new to HPC and experienced HPC users looking to adopt modern AI and MLOps workflows.

Keywords:

HAICORE, HPC, Scientific ML, MLOps, AI Workflows