M.Sc. Urvij Saroliya

Doctoral Candidate

Technical University of Munich
School of Computation, Information and Technology
Department of Computer Engineering
Informatics 10 – Chair of Computer Architecture and Parallel Systems (Prof. Dr. Martin Schulz)

Postal Address: Boltzmannstraße 3, 85748 Garching bei München
Room:  01.04.039
Tel: +49 (89) 289 - 17683
Email: urvij.saroliya@tum.de
Google Scholar

Research Interests

  • Computer Architecture
  • High Performance Computing (HPC)
  • Performance Analysis and Modeling
  • Performance Portability
  • Heterogeneous Architectures and Accelerators
  • HPC Resource Management
  • Power/Energy Aware HPC

Projects

Publications

  1. Urvij Saroliya, Eishi Arima, Dai Liu, Martin Schulz "Reinforcement Learning-driven Co-scheduling and Diverse Resource Assignments on NUMA Systems" In Proceedings of IEEE International Conference on Computer Design (ICCD), pp.xx--xx, November 2024
  2. Urvij Saroliya "Reinforcement Learning based Resource Management for HPC SystemsMaster’s Thesis in Informatics at Technical University of Munich (TUM), December 2023
  3. Urvij Saroliya, Eishi Arima, Dai Liu, Martin Schulz "Hierarchical Resource Partitioning on Modern GPUs: A Reinforcement Learning Approach" In Proceedings of IEEE International Conference on Cluster Computing (CLUSTER), pp. 185-196, November 2023
  4. Andreas Koch, Gabriel Dax, Michael Petry, Harvey Gomez, Amir Raoofy, Urvij Saroliya, Max Ghiglione, Gianluca Furano, Martin Werner, Carsten Trinitis, Martin Langer "Reference Implementations for Machine Learning Application Benchmark" In 2023 European Data Handling & Data Processing Conference (EDHPC), pp. 1-3. October 2023