Bengisu Elis

Raum: 01.04.037

Informatik 10 - Lehrstuhl für Rechnerarchitektur & Parallele Systeme (Prof. Schulz)

Technische Universität München

Boltzmannstr. 3

85748 Garching b. München

E-Mail: bengisu.elis@tum.de 

Tel.: +49 (89) 289 - 17676

Fax: +49 (89) 289 - 17662 

 

Publications

Conference and Journal Papers

  • A Mechanism to Generate Interception Based Tools for HPC Libraries. Bengisu Elis, David Boehme, Olga Pearce, and Martin Schulz. 2024. In Proceedings of International European Conference on Parallel and Distributed Computing (EuroPar '24).
  • Non-Blocking GPU-CPU Notifications to Enable More GPU-CPU Parallelism. Bengisu Elis, Olga Pearce, David Boehme, Jason Burmark, and Martin Schulz. 2024. In Proceedings of the International Conference on High Performance Computing in Asia-Pacific Region (HPCAsia '24). Association for Computing Machinery, New York, NY, USA, 1–11. ( HPCAsia'24 Best Paper Award Winner)
  • QMPI: A Next Generation MPI Profiling Interface. Bengisu Elis, Dai Yang, Olga Pearce, Kathryn Mohror, Martin Schulz.2020 Parallel Computing 
  • QMPI: A Next Generation MPI Profiling Interface. Bengisu Elis, Dai Yang, Martin Schulz. Proceeddings of the EuroMPI 2019 : 26th European MPI Users' Group Meeting.
  • The MPI Tool Interfaces: Past, Present, and Future — Capabilities and Prospects. 2019. Martin Schulz, Marc-André Hermanns, Michael Knobloch, Kathryn Mohror, Nathan T. Hjelm, Bengisu Elis, Karlo Kraljic, and Dai Yang. In book: Tools for High Performance Computing 2018 / 2019

Posters

  • MPI Tools the Easy Way. Bengisu Elis, Martin Ruefenacht, Olga Pearce, Kathryn Mohror, Martin Schulz, Anthony Skjellum. ISC20, June 2020.

 

Thesis

Master's Thesis : Design, Testing and Implementation of a new Profiling Interface for MPI

Teaching

Thesis

  • Stepan Vanecek (Master Thesis, 16.11.2020) Design, Implementation and Test of Efficient GPU to GPU Communication Methods
  • Matthias Kübrich (Bachelor Thesis , 22.10.2020)  Integration and test of RUST tool support for MPI