Thema der Dissertation:
Digital Twin for Energy-Aware High-Performance Computing Thema der Disputation:
Energy-aware High-Performance Computing: Challenges and Approaches
Digital Twin for Energy-Aware High-Performance Computing Thema der Disputation:
Energy-aware High-Performance Computing: Challenges and Approaches
Abstract: HPC clusters are becoming more and more powerful and require more energy. Widespread adoption of machine learning (ML) and the integration of accelerators into HPC clusters has intensified this trend. The first talk looks into this trend, presents methods to classify energy usage and introduces current possibilities for system operators to manage the energy use of their HPC system.
The second talk introduces a Digital Twin for energy-aware High-Performance computing. Digital Twins offer considerable advantages over conventional simulation methods and facilitate energy-aware scheduling and management of HPC clusters. The talk introduces different application scenarios and shows how the Digital Twin contributes to energy-aware HPC.
The second talk introduces a Digital Twin for energy-aware High-Performance computing. Digital Twins offer considerable advantages over conventional simulation methods and facilitate energy-aware scheduling and management of HPC clusters. The talk introduces different application scenarios and shows how the Digital Twin contributes to energy-aware HPC.
Zeit & Ort
26.01.2026 | 10:15
Seminarraum 005
(Fachbereich Mathematik und Informatik, Takustr.9, 14195 Berlin)