Skip to main content
Skip to article

Research Note

High Performance Computing

Zhenyu He · Jobs Stroustrup 1 min read

Proficiency

Proficient

Description

  • Large Eddy Simulation (LES) execution and debugging (DAM model), ~46k core-hours per run
  • Code extension: adding diagnostic output variables, new forcing logic, I/O cadence optimization
  • Compute resource planning: designing tractable sensitivity plans within compute budgets
  • Idealized aquaplanet simulations

Used In

Research-grade signature capabilities

PKU undergraduate HPC infra mastery (2020-2021)

  • Tianhe-II NSCC-GZ: 3 accounts ((undergraduate Tianhe-II account) primary + 2 more Walker variants); yhrun -p work -n 48 48-PE SLURM (rare for an undergraduate)
  • CAM3 aqua-planet 30-yr spin-up runs × 10 cases (D60-180 + Wonly + Conly + Aqua, each case a 30-year integration)
  • 27,882 MATLAB+NCL analysis lines + 5-stage HPC pipeline architecture (CAM3 aqua-planet → Tianhe-II runs → 25 cal_X MATLAB diagnostics → NCL stream function + ω analysis → Python figure polish)

UC Berkeley PhD HPC mastery (2022-2025)

  • Lawrencium LBNL cluster: 8 production runs × ~46k core-hours each ≈ 44 days HPC wall time engineering plan
  • DAM quasi-2D arrays Lx 20km-50000km scan (Phase 4.5 killer-figure convergence)
  • 115,363 Jupyter notebook lines / 62 ipynb / 3,793 code cells aggregate (PhD analysis code)

Cross-language HPC tooling

  • Fortran 77/90: source-code-level understanding of NCAR CAM3 v3.1 source tree (615 .F90 files) + DAM LES source (40+ modules)
  • Python/MATLAB/R/NCL/IDL/Mathematica: cross-language scientific computing (Caltech DSCOVR 5-language fluency)
  • remote workstation CUDA bootstrap (Caltech, Windows CUDA incompatibility → manual CUDA driver + Anaconda bootstrap)

Methodology concept pages