Ghanghoon “Will” Paik

HPC Machine Learning Performance Engineer & Infrastructure Strategist

Bridging the gap between academic research and industrial HPC solutions.

I am an HPC Machine Learning Performance Engineer with a PhD in Aerospace Engineering. With over 9 years of experience in high-performance computing, I focus on the intersection of hardware efficiency and real-world implementation.

Currently, I specialize in:

• Infrastructure Strategy: Optimizing TCO (Total Cost of Ownership) for high-performance computing
• System Architecture: Designing hybrid environments using Slurm and Kubernetes
• Performance Engineering: Tuning large-scale ML workloads for maximum efficiency

I maintain an Independent HPC Infrastructure Lab to experiment with enterprise-grade architectures and share my findings on my Blog (Launching in Jan 2026).

Professional Focus

• High-Performance Computing and Infrastructure I work with large-scale research computing environments, optimizing performance and designing solutions for computational challenges. My experience encompasses:

  • System Architecture & Management
    • Performance tuning and optimization for HPC clusters
    • Resource scheduling with Slurm and job optimization strategies
    • Distributed storage systems and high-speed interconnects
    • User support and computational workflow optimization
  • Infrastructure Automation and Deployment
    • Ansible playbooks for cluster configuration and management
    • Container environments with Singularity/Apptainer for reproducible research
    • System monitoring and performance metrics collection
    • Infrastructure as Code approaches for research computing
  • Emerging Technologies
    • Kubernetes orchestration for containerized HPC workloads
    • Hybrid cloud architectures for burst computing
    • Cost-benefit analysis of on-premises vs cloud solutions
    • Modern DevOps practices in research computing environments

• Computational Science Applications
  My computational background spans multiple domains:

  • Algorithm Development and Optimization
    • Parallel computing strategies for N-body simulations
    • Memory optimization for large-scale problems
    • Numerical methods for trajectory optimization
    • Performance profiling and bottleneck analysis
  • Research Computing Support
    • Helping researchers translate algorithms to HPC environments
    • Code optimization and parallelization strategies
    • Workflow design for computational pipelines
    • Best practices for scalable scientific computing

Current Projects

• Independent HPC Infrastructure Lab
  I maintain an independent HPC infrastructure lab where I develop and test various configurations relevant to modern research computing:

  • Small-Scale HPC Reference Implementations
    • Multi-node cluster with head node and compute nodes
    • NFS-based shared storage configurations
    • Ansible-automated deployment and management
    • Comparative analysis of different scheduler configurations
  • Cost Optimization Studies
    • TCO models for small-to-medium HPC deployments
    • Performance per dollar metrics across different architectures
    • Cloud vs on-premises decision frameworks
    • Resource utilization optimization strategies
  • Infrastructure Tools & Automation
    • Deployment scripts for common HPC configurations
    • Monitoring and alerting solutions
    • User management and quota systems
    • Backup and disaster recovery strategies

Technical Writing and Knowledge Sharing

I’m developing a series of technical articles focused on practical HPC implementation:

• Contributing to the Community
  • Active contributor to Singularity Hub
  • Technical documentation for HPC best practices
  • Open-source tools for cluster management

Background and Recognition

• Education
  • PhD in Aerospace Engineering - Penn State University
    • Dissertation: A Mission Planning Technique for Low-Thrust Synergetic Gravity-Assist Missions
    • Computational focus: High-performance algorithms for trajectory optimization
• Professional Experience
  • HPC Machine Learning Performance Engineer - Northeastern University Research Computing (2025-Present)
  • HPC Software Consultant - Penn State Institute for Computational and Data Sciences (2017-2024)
• Competitive Achievements
  • NIST First Responder UAS Indoor Challenge: First Responder’s Choice Award, Top 3 finish
  • Intelligent Ground Vehicle Competition: Team lead for autonomous systems
  • VFS Design-Build-Vertical Flight Competition: Recognition for innovative design
  • ESA Global Trajectory Optimization Competition: Multiple participations
• Technical Proficiencies
  • HPC Technologies: Slurm, PBS, OpenMPI, Intel MPI, module systems, job arrays, parallel filesystems
  • Infrastructure: Linux administration, NFS, Ansible, monitoring (Prometheus), virtualization (Proxmox), containerization (Singularity, Docker, Podman)
  • Development: Python, C/C++, MATLAB, shell scripting, parallel programming (MPI, OpenMP), CUDA
  • Cloud and Modern Tools: Kubernetes basics, AWS/Azure familiarity, GitLab CI/CD, infrastructure as code concepts

Get In Touch

I’m interested in discussing HPC infrastructure challenges, computational optimization strategies, and emerging technologies in research computing. Feel free to reach out for technical discussions or collaborations.