I work full-time in the High-Performance Computing (HPC) group at LRZ, specializing in high-level support, training, and extreme-scale computing. My expertise includes optimizing and parallelizing scientific application codes across diverse disciplines on novel HPC architectures. My main research interests focus on parallel computing, emerging programming languages, and new computing paradigms. Additionally, I am a DLI Certified Instructor, teaching fundamental CUDA and advanced multi-GPU scaling techniques for CUDA C++ applications across multiple nodes.
My main research interest in various aspects of parallel computing and new programming languages and paradigms.
I use my knowledge:
To support LRZ users in High Performance Computing
Low-level optimization, performance analysis and tuning on modern architectures (Link)
Maintaining softwares installation for Current & Upcoming processor architectures using package management tool Spack
(Link)
Porting and optomazing software on on Heterogeneous Supercomputers (e.g: CPU + GPUs
User support in using LRZ HPC resources for Machine Learning and Deep Learning(NVIDIA DGX systems@LRZ).
Benchmarking activity to evaluate the performance of high-end supercomputers