Collective communication
Developing high-performance collective communication on GPU platforms.
PhD student at the University of Maryland
HPC / ML SYS / SPARSE AND IRREGULAR PROBLEMS
LannieDaltonHough
I am a Computer Science PhD student at the University of Maryland in the Parallel Software and Systems Group. My research focuses on collective communication and high-performance systems for sparse and irregular problems.
Research areas
Sparse & irregular problems
Developing efficient representations and algorithms for workloads with irregular structure.
Performance analysis
Understanding bottlenecks and scaling behavior in distributed ML workloads.
Selected publications
SCTo appear2026
SpCCL: A Sparsity-Aware Collective Communication Library for GPU Platforms
Collective communication is essential to high-performance computing and machine learning workloads, and exploiting sparsity in message payloads can reduce communication overhead.
CAIS2026Understanding and Improving Communication Performance in Multi-node LLM Inference
As large language models continue to grow in size, distributed inference has become increasingly important.
SEA2025Elias-Fano Compression for Space-Efficient Rank and Select Structures
Cache-aware hierarchical rank/select data structure using Elias-Fano encoded L2 indices.
Background
About me
I am a CS PhD student at the University of Maryland in the Parallel Software and Systems Group. I earned my Bachelor of Science in Computer Science from Stetson University in 2021. After completing my undergraduate degree, I worked full-time as a software engineer at JPMorgan Chase & Co. from 2021–2022 and i2k Connect from 2022–2024.
At UMD, I am advised by Dr. Abhinav Bhatele. I research high-performance computing systems, with a current focus on collective communication and sparse and irregular problems.
Outside of work I enjoy LEGO, hiking, fantasy novels, video games, and retro computing.
Lannie Dalton HoughUniversity of MarylandEST // --:--:--