Tyler Allen
Assistant Professor, UNC Charlotte, Woodward 210E

Hello! I am Tyler Allen, Assistant Professor of Computer Science in the UNC Charlotte College of Computing and Informatics. My research focuses on enabling new capabilities for High Performance Computing (HPC) systems using architecture-focused applications, as well as making HPC systems more accessible through development and optimization of systems technology. Modern HPC systems are largely driven by graphics processing units (GPUs), and as such we focus on architecture-driven optimizations for GPU applications as well as effective methods for GPUs and their host systems to interact effectively. My goal is to transform performance optimization beyond simple metrics analysis into tangible, qualitative benefits for large-scale system users and applied applications developers.

I have several openings for Ph.D. students. If you are interested in these kinds of topics and wish to pursue a Ph.D., you can apply here and feel free to include my name in your application. I also highly encourage you to send me an email or visit my office on campus.


  • Heterogeneous Memory Systems
  • GPU Computing
  • Performance Optimization
  • Large-Scale Computing


Clemson University
2016 - 2022
Ph.D. Computer Science
Clemson University
2015 - 2018
M.S. Computer Science
Western Carolina University
2011 - 2015
B.S. Computer Science
Western Carolina University
2011 - 2015
B.S. Applied Mathematics

Recent Publications

See my google scholar for the latest list
“In-Depth Analyses of Unified Virtual Memory System for GPU Accelerated Computing”, 2021, In Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC ‘21)
Tyler Allen , Rong Ge
FULL-W2V: Fully Exploiting Data Reuse for W2V on GPU-accelerated Systems, 2021, ACM International Conference on Supercomputing (ICS-2021)
Thomas Randall , Tyler Allen , Rong Ge
“Demystifying GPU UVM Cost with Deep Runtime and Workload Analysis”, 2021, IEEE International Parallel and Distributed Processing Symposium (IPDPS)
Tyler Allen , Rong Ge