About us
The University of Luxembourg is an international research university with a distinctly multilingual and interdisciplinary character.
The Faculty of Science, Technology and Medicine (FSTM) at the University of Luxembourg contributes multidisciplinary expertise in the fields of Mathematics, Physics, Engineering, Computer Science, Life Sciences and Medicine.
Through its dual mission of teaching and research, the FSTM seeks to generate and disseminate knowledge and train new generations of responsible citizens in order to better understand, explain and advance society and the environment we live in.
The HPC platform team at the University of Luxembourg runs the university's large academic high-performance computing facility, a research infrastructure for public research in Luxembourg that also serves as an edge-access solution to the EuroHPC Luxembourg supercomputer. The centre was founded to meet growing demand for computational resources and acts as a central hub for research and development. Its mission is to drive scientific discovery, enable high-impact research, and create opportunities for collaboration across academia, industry, and government.
The team provides state-of-the-art facilities and expert guidance so that users from many fields can use advanced computing power in their research and innovation work. It also takes an active part in international partnerships, contributing to European and global HPC initiatives. The facility's two clusters are Iris and Aion; Aion was inaugurated by the university as a supercomputer and has recently been improved. Beyond running the infrastructure, the team shares news on how HPC can help with problems such as large-scale manufacturing optimization, and on access to cloud computing resources such as Amazon's through project calls.
Your role
This position is funded within the HPC Platform team of the University of Luxembourg and combines research with the practical needs of a production university HPC service. It addresses efficiency across the whole stack, and we welcome candidates whose strength lies in either of two complementary directions, or in both:
- Infrastructure and hardware design: low-energy cluster design, including direct liquid cooling (DLC), immersion cooling and low-cost or self-built cooling solutions, the use of mid-range and lower-end GPUs and accelerators, power capping, and heat reuse, evaluated for performance, reliability, operating effort and total cost
- Software and benchmarking: a uniform, reproducible benchmarking approach at the level of the numerical-library interfaces that applications depend on (BLAS, LAPACK, FFT, sparse routines and AI operator libraries). Mainstream vendors provide tuned implementations of these, while emerging and low-power platforms often cover them only partially
Both directions feed the same cost, power and performance model, so that design choices at the facility level (for example the cooling approach or the GPU tier) can be compared with software-level choices (for example precision, library implementation or scheduling policy) on a common basis.
Over the course of the contract, the expected outcomes are:
- A validated cost, power and performance model (including TFLOPS per watt, PUE and total cost of ownership over the system lifetime, covering energy, cooling, hosting and staff effort) for university-scale hardware options across budget tiers up to €10M
- A comparative evaluation of low-energy cluster designs, including air cooling, direct liquid cooling and immersion cooling, and of GPU tiers (from lower-end and consumer-class to data-centre accelerators), with measured power, thermal behaviour, reliability and operating effort, and practical build guides for low-cost solutions
- An analysis of the role of university clusters in the wider HPC ecosystem, comparing local ownership with national allocations, cloud bursting and inter-university federation (for example within the UniGR), and identifying which workloads belong where
- A reproducible benchmarking suite, built on open-source software stacks, based on a functional API of numerical-library routines and synthetic tests that represent typical application workloads, with published results for vendor-provided implementations on commodity, heterogeneous and emerging hardware (covering performance, energy and numerical accuracy)
- Reference cluster designs at several budget tiers (for example around €250k, €1M, €3M and €10M), with measured performance, power and TCO, and guidance on sizing, cooling, procurement and phased upgrades
Your profile
- PhD in Computer Science, Electrical Engineering, Software Engineering, Mechanical or Energy Engineering, or a related field
- Research or professional experience in High-Performance Computing (HPC), systems engineering, or distributed systems. Your main strength may lie in infrastructure and hardware design , e,g, cooling, power, system integration, or in software and benchmarking (for example numerical libraries, performance engineering, ML optimisation
- Experience with computing clusters or data-centre infrastructure, e.g. SLURM, Kubernetes, Docker or Apptainer, or liquid cooling and power measurement
- Familiarity with benchmarking and performance profiling, e.g. HPL, HPCG, MLPerf or vendor profilers
- Interest in sustainability and energy-efficient computing, with the ability to carry out energy and cost analysis
- A hands-on approach and a taste for solving practical problems with limited resources
- Publications in relevant fields, e.g. HPC, systems, or AI/ML systems, are expected Contributions to open-source projects are an advantage
- Excellent command of English; good working knowledge of French or German is a plus
Advantages (not required):
- Experience with machine learning optimisation, e.g. uantisation, pruning, distributed training or inference serving
- Familiarity with numerical libraries, e.g. LAS/LAPACK, cuBLAS, rocBLAS, oneMKL
- Experience operating shared research clusters, ideally in a university setting
- Familiarity with TCO or lifecycle cost analysis
We offer
- A modern, dynamic university with a personal and inclusive atmosphere. Multilingual and international character. Staff coming from more than 90 countries. Member of The Guild of European Research Intensive Universities
- An exceptional research environment, supported by skilled staff and high-quality equipment. Strong links to professional sectors and the Luxembourg labour market. A unique urban campus with excellent infrastructure
- A partner for society and industry. Cooperation with European institutions, innovative companies, the Financial Centre and a wide range of non-academic partners including ministries, local governments, associations, and NGOs
How to apply
Applications in PDF format should include:
- Curriculum Vitae
- Cover letter detailing your motivation for applying to the advertised research topic and/or project, including how your background, interests, and career goals align with its objective
- PhD diploma or a letter/information indicating the expected defense date
- Transcript of all modules and results from university-level courses taken
- List of publications
- Names, contact information and current position of 1-2 referees who can, upon request, provide an assessment of the application. Please indicate their relationship to you
Early application is highly encouraged, as the applications will be processed upon reception. Please apply ONLINE formally through the HR system. Applications by Email will not be considered.
All qualified individuals are encouraged to apply. In line with our values, the University of Luxembourg promotes an inclusive culture. We encourage applications from individuals of all backgrounds and are dedicated to upholding equality and respect for our employees and students.
General information:
- Contract Type: Fixed Term Contract 24 Month
- Work Hours: Full Time 40.0 Hours per Week
- Location: Belval Campus
- Internal Title: Postdoctoral researcher
- Job Reference: UOL08593
The yearly gross salary for every Postdoctoral Researcher at the UL is EUR 87306 (full time).
