Biography
I am a Graduate Research Assistant at The University of Toledo, pursuing my Master’s in Computer Science. My research focuses on High-Performance Computing (HPC) and Stochastic Modeling & Optimization. I am passionate about developing innovative system software solutions for HPC systems to enhance their efficiency, reliability, and cost-effectiveness.
My current research involves optimizing scientific workflows on edge computing systems and developing autotuning frameworks for HPC applications. I have experience in parallel computing, distributed systems, and machine learning infrastructure. I am particularly interested in the intersection of HPC and artificial intelligence, exploring ways to leverage high-performance computing resources for advanced AI applications.
Before joining the University of Toledo, I completed my Bachelor’s in Electrical Engineering from Bangladesh.
- High Performance Computing (HPC)
- Stochastic Modeling & Optimization
- Parallel Computing
- Machine Learning Systems
- Cloud Computing
MSc in Computer Science, 2025 (Expected)
The University of Toledo
BSc in Electrical and Electronic Engineering, 2019
Chittagong University of Engineering & Technology
Experience
Responsibilities included:
- Set up CouchDB, Chords, and Streampipes on ACCESS Jetstreams for community weather data storage
- Built data orchestrator for efficient data routing, achieving 30% transmission efficiency gain
- Reduced deployment costs by 25% for communities implementing the project
Key achievements:
- Designed private LoRa network for 6+ data types with Raspberry Pi gateways and central server
- Improved wind forecasting with edge-ML, achieving 20% accuracy gain on Raspberry Pi
- Image analysis with TensorFlow, 25% accuracy gain, 3x faster training, and 95% precision on 10,000+ images
Research focus:
- Contributed to an NSF-funded project to design HPEE, a new auto-tuning algorithm for optimizing HPC applications on edge devices
- Achieved 2.5X more efficiency with HPEE than benchmark methods
- Improve High-performance computing (HPC) systems through stochastic modeling and optimization
Publications
2024
“HPC Application Parameter Autotuning on Edge Devices: A Bandit Learning Approach”
Abrar Hossain, Abdel-Hameed Badawy, Mohammad Atiqul Islam, Tapasya Patki, Kishwar Ahmed
IEEE International Conference on High Performance Computing, Data, and Analytics (HiPC'24)“Scalable HPC Job Scheduling and Resource Management in SST”
Abubeker Abdurahman, Abrar Hossain, Kevin A Brown, Kazutomo Yoshii, Kishwar Ahmed
Winter Simulation Conference (WSC'24)“Environmental Data Sensing and Monitoring System Using Community-based Private LoRa Network”
Abrar Hossain, Keith Maull, Agbeli Ameko American Geophysical Union (AGU'24)
2023
- “Automating HPC Model Selection on Edge Devices”
Abrar Hossain, Kishwar Ahmed
International Conference for High Performance Computing, Networking, Storage and Analysis (SC'23)
Recorded Talks
Internship Presentations
Environmental Data Sensing and Monitoring System Using Community-based Private LoRa Network
Venue: NCAR Mesa Lab, Boulder, CO
Date: August 2024
Abstract: Talk on my summer research with Dr Keith Maull and Agbeli Ameko