Signal-Processing and Data-Compression on Beaglebone Black used as embedded HPC-performance-monitoring device
In recent years, energy-efficiency is becoming a key challenge in the High Performance Computing (HPC) domain. Indeed, while the demand for more powerful supercomputers is constantly increasing, integrated computing architectures are facing power/thermal challenges that are limiting the performance benefits of technology scaling. However, even if energy-efficiency is of primary interest, such infrastructures have a limited introspection on the measurement of power and energy dissipated at run-time. Aiming at bridging this gap, we designed and developed at IIS, in collaboration with some Commercial Partners, a scalable and highly accurate power-consumption monitoring framework, based on a state-of-the-art embedded monitoring device (namely a Beaglebone Black) and a power-sensing-board placed at the node power-supply.
Goal of this Project → Evaluation of different compression and spectral information extraction strategies (based on open-source ARM-Linux loss-less implementations when available) for the best performance on our application-scenario.
- Good knowledge on signal processing and C programming.
- Experience using laboratory instrumentation - signal generators, oscilloscopes, Matlab, etc.
- Experience on microcontrollers would be an asset.
- 40% Theory
- 40% Implementation (C programming for embedded systems)
- 20% Testing
Detailed Task Description