Jesper Larsson Träff

Prof. Jesper Larsson Träff is a Full Professor and Head of the Research Unit in Parallel Computing at Technische Universität Wien (TU Wien). His academic role is centered within the Department of Computer Engineering. He holds an MSc and PhD, and has contributed extensively to the field of parallel computing through his research and leadership in international projects.

His research focuses on parallel algorithms, scheduling, and optimization of the Message Passing Interface (MPI), with emphasis on high-performance computing (HPC) systems and distributed architectures. Key projects include the Process Mapping initiative (2019–2024), Autotune (2021–2025), and contributions to Exascale Programming Models. He has also led efforts in evaluating MPI performance tools and their reproducibility challenges.

• Research Areas: Parallel Algorithms, MPI Optimization, HPC, Process Mapping, Collective Communication, and Memory Models.
• Notable Achievements: Awarded the Best Paper at EuroMPI 2014 and recognized by the Innovation Radar for his work on PGAS-based MPI interoperability in 2018.
• Teaching: Teaches courses such as Advanced Multiprocessor Programming, Bachelor/Master Thesis supervision, and seminars in Computer Engineering and Theoretical Computer Science.

Prof. Träff’s advising includes students researching topics like lock-free data structures, MPI datatype optimization, and task scheduling. His grants span Austrian and EU funding bodies, addressing challenges in exascale computing and reproducible experimental research. He is affiliated with the Vienna Mapping and Sparse Quadratic Assignment (Vienna Mapping) project and actively contributes to international workshops and conferences.