Presenter

Srinath Vadlamani is a principal performance engineer at HPE’s Center of Excellence for Lawrence Livermore National Laboratory (LLNL), specializing in high performance computing (HPC) and performance portability across heterogeneous architectures. He has over 15 years of experience optimizing mission-critical scientific applications on leadership-class systems, including AMD Instinct MI300A APU architectures for El Capitan.

Srinath's expertise spans MPI, OpenMP, HIP, CUDA, and OpenACC, with deep architectural knowledge of x86, Arm, and GPU platforms. He is proficient in ROCm and CUDA profiling, GPU occupancy tuning, register pressure analysis, and power-aware scaling.

Previously, Srinath held roles at Arm Inc., ParaTools, NCAR, and Tech-X Corp., leading HPC code modernization, reproducible benchmarking, and accelerator integration. An active contributor to Spack, ReFrame, and the INCITS Fortran Standards Committee, he earned his PhD in applied mathematics from the University of Colorado Boulder, where his dissertation unified particle-in-cell and continuum methods for modeling plasma instabilities.