Aktuelles

Author: Urvij Saroliya; Advisors: Eishi Arima, Martin Kronbichler; Doctoral Supervisor: Martin Schulz [weiterlesen]

Authors: D. Huber, S. Iserte, M. Schreiber, P.-F. Dutot, O. Richard, A. Peña, K. Gaddameedi, T. Neckel, H.-J. Bungartz, M. Schulz [weiterlesen]

Accepted paper at ProTools'25 (SC Conference Workshop) titled MT4G: A Tool for Reliable Auto-Discovery of NVIDIA and AMD GPU Compute and Memory Topologies by Stepan Vanecek et al. [weiterlesen]

CAPS recently had the pleasure of visiting the German Aerospace Center (DLR) in Oberpfaffenhofen, one of Germany’s largest and most advanced research hubs. Thanks to our colleagues Maron and Cedric—external PhDs at CAPS working at DLR—we got an inside look at how high-performance computing meets… [weiterlesen]

Authors: Mohak Chadha, Pulkit Khera, Jianfeng Gu, Osama Abboud (Huawei MRC), and Michael Gerndt Abstract: Federated Learning (FL) is an emerging… [weiterlesen]

The paper "Systematic Analysis of DDS Implementations", a result of a TUM collaboration with Siemens, was accepted at the Middleware 2023 conference. [weiterlesen]

Authors: Xiaorang Guo, Kun Qin and Martin Schulz [weiterlesen]

Authors: Jonas Winklmann, Martin Schulz and Dimitrios Tsevas [weiterlesen]

The paper titled "Hierarchical Resource Partitioning on Modern GPUs: A Reinforcement Learning Approach" got accepted at IEEE Cluster 2023 [weiterlesen]