Embedded Systems

Performance Modeling of the NVIDIA Deep-Learning Accelerator (NVDLA) using Performance Representatives

Bachelor’s Thesis / Master’s Thesis / Student Research Project


This student project’s goal is to perform benchmarks on the Nvidia NVDLA Machine Learning Accelerator and use those measurements to create a statistical performance estimator using the Performance Representatives (PR) approach. This model should then be compared to existing analytical models, like AMAIX, and other performance estimation approaches, like ACADL/AIDG.

NVDLA Accelerator



  • Python
  • Linux
  • Successfully atteded the lecture “Grundlagen der Rechnerarchitektur” and/or “Parallele Rechnerarchitekturen” and/or “Modellierung und Analyse Eingebetteter Systeme” (optional)


Jung, Alexander

Lübeck, Konstantin

Bringmann, Oliver