Hello!

I'm Alex Villa, a CSE major at UC Merced, passionate about MPI, GPU offloading, and data parallelism.

About Me

I'm an undergraduate Computer Science & Engineering student at UC Merced (Class of 2026). My research delves into collective communication (MPI, NCCL), GPU offloading, and scalable parallel frameworks. Driven by curiosity, I thrive on crafting efficient HPC solutions to tackle real-world scientific challenges.

Research Interests

My research focuses on high-performance computing (HPC) techniques, particularly in the areas of:

I'm passionate about leveraging these technologies to solve complex scientific problems and improve computational efficiency.

Experience

Los Alamos National Laboratory

Summer Research Intern, Computational Physics Division

Jun 2024 – Aug 2024
  • Designed & implemented C++ pointer support in FleCSI (C++17/20/23), enabling direct task parameterization of complex data.
  • Optimized distributed CPU/GPU memory access & validation for asynchronous task pipelines in scientific workloads.
  • Leveraged Legion & NVIDIA HPC SDK to enhance memory safety and performance in computational physics simulations.
View Project

UC Merced

Undergraduate Research Assistant, PADSYS Lab

Oct 2023 – Present
  • Presented MPI & distributed DL proxy benchmarks (CabanaMD, miniMD) in in-depth research talks.
  • Analyzed Horovod collective patterns (ring vs butterfly Allreduce) & compared MPICH, NVIDIA NCCL, Horovod implementations.
  • Prototyped optimizations to accelerate DL model parallelism and evaluate communication bottlenecks.

UC Merced

Student Technology Consultant, Cyberinfrastructure Team

Oct 2022 – Present
  • Support HPC users: job scheduling, environment modules, debugging on campus cluster.
  • Authored & maintained comprehensive HPC user documentation & tutorials.
  • Led workshops on SLURM, MPI, containerization (Singularity/Apptainer).
Docs Site

Preprints and Publications

Awards & Honors

Talks & Presentations

Education

Get in Touch

📧 avilla@lanl.gov

💼 LinkedIn | GitHub