Related Experiment Video
Updated: Jul 30, 2026
09:53
Micropatterning Transmission Electron Microscopy Grids to Direct Cell Positioning within Whole-Cell Cryo-Electron Tomography Workflows
Published on: September 13, 2021
Simulation of emission tomography using grid middleware for distributed computing.
M G Thomason1, R F Longton, J Gregor
1Department of Computer Science, University of Tennessee, 203 Claxton Complex, Knoxville, TN 37996, USA. thomason@cs.utk.edu
Computer Methods and Programs in Biomedicine
|July 22, 2004
Summary
Parallelizing SimSET Monte Carlo simulations using NetSolve grid middleware achieves linear speed-up. A novel scheduler optimizes performance across heterogeneous computing environments for emission tomography.
Area of Science:
- Medical Imaging
- Computational Science
- Distributed Computing
Background:
- SimSET is Monte Carlo simulation software crucial for emission tomography.
- Efficient parallel execution is needed to handle complex simulations.
Purpose of the Study:
- To describe a scheme for parallelizing SimSET using NetSolve grid middleware.
- To implement a client-side scheduler for optimizing performance in heterogeneous environments.
Main Methods:
- Utilized NetSolve (version 1.4.1) as client-server grid middleware.
- Developed a client-side scheduler to partition decay events based on processor compute-speeds and workloads.
- Made minimal modifications to SimSET code (version 2.6.2.3) to ensure total decay event count.
Main Results:
- Achieved essentially linear speed-up with the number of processors when considering compute-speeds and workloads.
- Demonstrated effective parallel execution in a heterogeneous computing environment.
- Reported empirical results for simulating a clinical-quality lung perfusion study.
Conclusions:
- The NetSolve-based parallelization scheme is effective for SimSET.
- The client-side scheduler enhances performance by adapting to diverse computing resources.
- This approach enables efficient distributed computation for emission tomography simulations.
