Mohak Chadha

Design and development of parallel benchmarks for non-intrusive online tracing on embedded multicore systems (ARM & PowerPC) as part of the COEMS and the ARAMIS projects. Development of a web interface for real time visualization and analysis of online traces.

I worked under the supervision of Prof. Dr. Michael Gerndt. The work involved the development of DVFS and UFS plugins for autotuning HPC applications at a region level for Energy-related Tuning objectives as part of the Runtime Exploitation of Application Dynamism for Energy Efficient eXascale computing (READEX) project. The work was published in the 33rd IEEE International Parallel and Distributed Processing Symposium (IPDPS).

I was part of the automotive software team. I worked on the design and development of a real-time robust computer vision based lane detection system for embedded platforms with Nvida-Gpus.

I worked under the supervision of Prof. Dr. Wolfgang E. Nagel. I worked on the development of a modeling methodology which utilizes statistical analysis and regression techniques to formulate accurate, real-time power models for HPC systems. The work was published in 31st IEEE International Parallel and Distributed Processing Symposium Workshops.

I worked under the supervision of Dr. Santonu Sarka where we designed and developed a unified power and energy measurement application programming interface for HPC developers. The work was published at 35th International Performance Computing and Communications Conference 2016 held in Las Vegas, USA.

Computer Architecture, Parallel and HIgh Performance Computing, Machine Learning