Scientific Research Assistant SEANERGYS Job-Misconfiguration

As HPC systems are growing larger and the demand for AI keeps increasing, energy efficiency becomes ever more important for sustainable HPC operations. LRZ has a long track record of R&D activities in this field and its HPC operations have been driving improvements in energy efficiency for more than 15 years.

LRZ will be contributing to the European research project SEANERGYS (scheduled to start in Q1/2025) that is jointly funded by the EuroHPC JU and BMBF to build an integrated European software solution that optimises HPC resource utilisation and reduces the energy used for real-world workload mixes. It therefore improves the throughput of HPC systems, generating more R&D results for a given energy budget. The solution consists of a comprehensive monitoring infrastructure (CMI), an Artificial Intelligence data analytics system (AIDAS), and a dynamic scheduling and resource management system (DSRM). LRZ’s contributions for SEANERGYS will focus on the CMI and AIDAS sub-components. In CMI we will contribute to the development of a state-of-the-art monitoring solution for HPC operations that encompasses all relevant domains: system hardware, system software, applications, and building infrastructure. In AIDAS we will develop (AI-based) analysis methods to characterize HPC workloads already at submission to the job scheduler to facilitate their energy-efficient execution. We will also develop tools to detect mis-configurations of jobs and to devise optimal configurations. For this, LRZ is looking for a

The SEANERGYS project is a collaboration of LRZ with 14 European partners: Barcelona Supercomputing Center, CINECA, E4, Eviden, IT4innovation, Jülich Supercomputing Centre, LuxProvide, National Technical University of Athens, ParTec, Ryax Technologies, Technical University of Dresden, Technical University of Munich, University of Bologna, and University of Grenoble.

• Develop methods to identify misconfigurations of HPC jobs (thread pinning, use of wrong libraries, wrong thread/process ratio)
• Collaborate with project partners from other European HPC sites, universities and companies on the design and development of the solution, employing project-wide CI/CD. 
• Roll-out and support of the newly developed solution on LRZ production systems.
• Disseminate your findings in peer-reviewed publications and present them at national and international conferences.

• Master or PhD in Computer Science or similar field
• Good knowledge and experience in C, C++, Python
• Interest in experimenting with new ideas and work with other people to come up with novel and unconventional solutions
• Ability to share results clearly in publications and engagement in enlarging your network in the HPC community at conferences
• Interest in supporting the “Future Computing” research team at LRZ working on the evaluation of new technologies

• Experience with usage of HPC resources (batch scheduler, job scripts, MPI)
• Experience with HPC performance analysis
• Experience with HPC operations
• Published peer-reviewed scientific papers in HPC
• Interest in advising student work

Are you looking for a multifaceted and intellectually stimulating position in a dynamic, cooperative and innovative work environment? Then LRZ is the place to be for you! Here at LRZ a collegial, appreciative work environment meets an international crowd of experts who work together to advance IT services for ground-breaking research. We offer flexible work schemes for an optimal work-life balance. Our staff values their creative leeway. As an institute of the Bavarian Academy of Science and Humanities we offer all the benefits of public service. And of course, no wishes remain unfulfilled at the LRZ in terms of technical equipment. We share experiences, constantly review and improve our processes, and are proud that our service-quality and data-security are regularly certified and rated highly.

We actively promote diversity and welcome applications from talented individuals, regardless of cultural background, nationality, ethnicity, gender and sexual identity, physical abilities, religion and age. We give priority to applications from people with disabilities who are equally qualified (SGB IX).   

We are looking forward to receiving your complete application documents (including cover letter, CV and certificates) in a PDF file (other file types are not accepted) by latest 13.12.2024.

Subject: Job-Misconfiguration (2024/50)

Are you unsure whether the job suits you or you suit us? Or do you still have questions about this position? Our colleagues will be happy to answer all your questions. 

This job does not fit? Then take a look at https://www.lrz.de/wir/stellen/ or send us an unsolicited application!