Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Parallel global optimization with the particle swarm algorithm.

J F Schutte1, J A Reinbolt, B J Fregly

  • 1Department of Mechanical and Aerospace Engineering, University of Florida, Gainesville, FL, U.S.A.

International Journal for Numerical Methods in Engineering
|September 25, 2007
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Synergy Feedback Control Predicts Walking Across Multiple Cycles.

bioRxiv : the preprint server for biology·2026
Same author

The Neuromusculoskeletal Modeling Pipeline: MATLAB-based Model Personalization and Treatment Optimization Functionality for OpenSim.

bioRxiv : the preprint server for biology·2024
Same author

The effect of plate design, bridging span, and fracture healing on the performance of high tibial osteotomy plates: An experimental and finite element study.

Bone & joint research·2019
Same author

Mineralisation of filamentous cyanobacteria in Lake Thetis stromatolites, Western Australia.

Geobiology·2018
Same author

Optimizing whole-body kinematics to minimize valgus knee loading during sidestepping: implications for ACL injury risk.

Journal of biomechanics·2012
Same author

[Responsible antibiotic use].

Tijdschrift voor diergeneeskunde·2009

Parallel particle swarm optimization (PSO) offers significant speedups for load-balanced engineering problems. However, load imbalance in complex biomechanical systems reduces efficiency, highlighting the need for asynchronous implementations and larger particle populations.

Area of Science:

  • Computational Science
  • Engineering Optimization
  • Parallel Computing

Background:

  • Engineering optimization problems often require extensive computation, leading to long solution times.
  • Particle Swarm Optimization (PSO) is a popular global search method for these problems.
  • Coarse-grained parallelization can enhance computational throughput and global search capabilities.

Purpose of the Study:

  • To detail the coarse-grained parallelization of the Particle Swarm Optimization (PSO) algorithm.
  • To evaluate the performance of Parallel PSO on both computationally cheap and expensive optimization problems.
  • To identify factors affecting parallel performance and suggest improvements.

Main Methods:

  • Parallelization of the PSO algorithm using a coarse-grained approach.

Related Experiment Videos

  • Evaluation on large-scale analytical test problems (computationally cheap) and medium-scale biomechanical system identification problems (computationally expensive).
  • Analysis of speedup and parallel efficiency under varying load-balanced and load-imbalanced conditions.

    • Main Results:

    • Near-ideal speedup and over 95% parallel efficiency for load-balanced analytical problems on up to 32 nodes.
    • Diminishing speedup and linearly decreasing parallel efficiency for load-imbalanced biomechanical problems.
    • Synchronization requirements (waiting for slowest fitness evaluation) were identified as a key performance bottleneck.
    • A single large population (128 particles) showed better convergence than multiple smaller sub-populations for analytical problems.

    Conclusions:

    • Parallel PSO demonstrates excellent performance in load-balanced scenarios.
    • Asynchronous implementations are crucial for improving performance on real-world, load-imbalanced problems.
    • Larger population sizes are recommended when utilizing multiple processors for enhanced convergence.