🎓 Student at University of Electronic Science and Technology of China (UESTC)
📍 Based in China
💡 Focusing on GPU Computing, High-Performance Computing, Industrial Software & Databases
- 🚀 GPU-accelerated fluid simulation software kernels
- 📊 GPU-accelerated SpMV operators research
- 🗄️ GPU-accelerated vectorized query database kernels
- ⚡ High-performance computing solutions with CUDA
Primary Languages: C++ | CUDA | Python
Focus Areas:
- GPU Computing & CUDA Programming
- High-Performance Computing (HPC)
- Parallel & Concurrent Programming
Header-only CUDA runtime library for automatic dependency-aware kernel scheduling based on memory access semantics.
CUDA concurrency library that simplifies parallel programming by automating concurrency flow management with C++-style wrappers.
Lightweight deep learning library in C++ with CUDA C for efficient GPU-accelerated model training and inference.
Code for our paper submitted to ICDE 2027 — T-BLAEQ: A Tensor-Based Multigrid Index for GPU-Accelerated Multi-dimensional Query Processing. T-BLAEQ introduces a novel tensor-based multigrid indexing structure that leverages GPU parallelism to significantly accelerate multi-dimensional range and similarity queries, targeting high-throughput analytical workloads in modern database systems.
- 📧 Email: mgepahmge@outlook.com
- 🌐 Website: mgepahmge.dev
- 💬 Open to collaborations on GPU computing, HPC, and database projects
"Making parallel computing simpler and faster"