Related Experiment Video
Updated: Mar 16, 2026
05:37
Rapid in-silico Battery Electrolyte Electrochemical Reaction Generation using 3T-VASP Multi-Scale Energy Minimization
Published on: August 22, 2025
754
Evaluation of Emerging Energy-Efficient Heterogeneous Computing Platforms for Biomolecular and Cellular Simulation
John E Stone1, Michael J Hallock2, James C Phillips1
1Beckman Institute, University of Illinois at Urbana-Champaign, Urbana, IL, 61801, USA.
IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum : [Proceedings]. IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum
|August 13, 2016
Summary
Researchers are optimizing computational biology applications for mobile and exascale systems. This involves adapting software for heterogeneous computing platforms to enhance energy efficiency and performance for molecular modeling tasks.
Area of Science:
- Computational Biology and Bioinformatics
- High-Performance Computing (HPC)
- Mobile Computing and Graphics Processing Units (GPUs)
Background:
- Scientific advances in computational biology rely on increasing computational power for detailed cellular process simulations.
- Developing energy-efficient exascale supercomputers requires new hardware and applications to improve simulation, analysis, and visualization performance.
- Mobile platforms now offer sufficient power for interactive molecular visualization, enabling new collaboration and immersive viewing opportunities.
Purpose of the Study:
- To adapt biomolecular simulation and analysis applications for emerging heterogeneous computing platforms.
- To evaluate the performance and energy efficiency of these applications on new hardware architectures.
- To identify bottlenecks and suggest improvements for molecular modeling on mobile devices and future exascale computers.
Main Methods:
- Adapted biomolecular simulation and analysis applications for heterogeneous platforms combining CPUs and GPUs.
- Utilized low-cost power monitoring instrumentation to measure energy consumption of CPU algorithms and GPU kernels.
- Compared performance and energy efficiency against traditional computing platforms.
Main Results:
- Demonstrated early experiences adapting applications for power-efficient heterogeneous systems.
- Identified specific hardware and algorithmic bottlenecks hindering usability on emerging platforms.
- Provided comparative performance and energy efficiency data between traditional and emerging platforms.
Conclusions:
- Heterogeneous computing platforms offer potential for energy-efficient molecular modeling.
- Further hardware and algorithmic optimizations are necessary to fully leverage these platforms for mobile and exascale computing.
- This work lays the groundwork for future advancements in computational biology on next-generation systems.
