A custom C/OpenMP implementation of a Dynamic Affinity Scheduling algorithm, optimizing computational load balancing and minimizing lock contention in High-Performance Computing (HPC) environments.

Demonstrating HPC performance optimization. Took a naive N-body MD simulation from 872s → 27.4s on AMD EPYC 7742 (Zen 2) using cache-aware programming, loop fusion, and compiler tuning. No parallel code, pure single-core optimization.

LAMMPS course for the CCC-ParaSolS workshop with updated template