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

GPU-FS-kNN: a software tool for fast and scalable kNN computation using GPUs.

Ahmed Shamsul Arefin1, Carlos Riveros, Regina Berretta

  • 1Centre for Bioinformatics, Biomarker Discovery and Information-Based Medicine, The University of Newcastle, Callaghan, New South Wales, Australia.

Plos One
|September 1, 2012
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Fast Fourier Transform01:10

Fast Fourier Transform

The Fast Fourier Transform (FFT) is a computational algorithm designed to compute the Discrete Fourier Transform (DFT) efficiently. By breaking down the calculations into smaller, manageable sections, the FFT significantly reduces the computational complexity involved. Direct computation of an N-point DFT requires N2 complex multiplications, whereas the FFT algorithm needs only (N/2)log⁡2N multiplications, offering a much faster performance.
The computational efficiency of the FFT becomes...

You might also read

Related Articles

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

Sort by
Same author

First-Line Enfortumab Vedotin Plus Pembrolizumab vs Gemcitabine Plus Cisplatin for Metastatic Urothelial Carcinoma.

JAMA network open·2026
Same author

Comparative analysis of genitourinary toxicity following bladder outlet obstruction procedures in patients before versus after radiotherapy for localized prostate cancer: a retrospective cohort study using the TriNetX database.

Prostate cancer and prostatic diseases·2026
Same author

Associations between erectile dysfunction, low testosterone, and cardiometabolic risk: an age stratified, propensity-matched cohort study.

International journal of impotence research·2026
Same author

Addition of immune checkpoint inhibitors to intravesical BCG for high-risk BCG-naïve non-muscle invasive bladder cancer: Systematic review and meta-analysis.

BJUI compass·2026
Same author

Coverage, Traits, and Geographic Distribution of Online Surgeon Reviews: Large-Scale Cross-Sectional Analysis.

JMIR formative research·2026
Same author

Real-world functional and oncologic outcomes of prostate gland ablation vs. standard of care therapies for localized prostate: A retrospective cohort study using the TriNetX database.

Urologic oncology·2026

This study introduces a fast and scalable GPU-based method for k-Nearest Neighbour (kNN) search, crucial for analyzing large biological datasets. The GPU-FS-kNN tool significantly accelerates computations, offering a powerful solution for bioinformatics and machine learning challenges.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Machine Learning

Background:

  • High-throughput techniques generate massive biological datasets, demanding significant computational power.
  • Analyzing large biological networks presents computational challenges, often requiring supercomputers.
  • General Purpose computation on Graphics Processing Units (GPGPU) offers a cost-effective solution but faces scalability issues due to memory limitations.

Purpose of the Study:

  • To develop an efficient parallel formulation for the k-Nearest Neighbour (kNN) search problem.
  • To address the computational intensity and performance degradation of kNN with large datasets.
  • To create a fast and scalable solution for analyzing large-scale biological networks.

Main Methods:

  • Developed an efficient parallel formulation for kNN search.

Related Experiment Videos

  • Implemented a software tool, GPU-FS-kNN, utilizing CUDA for Graphics Processing Units (GPUs).
  • The approach emphasizes data and computational parallelism with partitioning for scalability.

    • Main Results:

    • The GPU-FS-kNN approach demonstrates significant speed-ups (50-60 times) compared to CPU implementations.
    • Achieved fast and scalable performance on large-scale instances, including a breast microarray dataset.
    • The method is adaptable to various GPU architectures.

    Conclusions:

    • The GPU-based Fast and Scalable k-Nearest Neighbour search technique (GPU-FS-kNN) significantly enhances nearest neighbour computation performance.
    • Provides a viable and efficient solution for analyzing large-scale biological networks.
    • Source code and software are available under GNU Public License (GPL).