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

Comparing machine-independent versus machine-specific parallelization of a software platform for biological sequence

P L Miller1, P M Nadkarni, W R Pearson

  • 1Department of Anesthesiology, Yale University School of Medicine, New Haven, CT 06510.

Computer Applications in the Biosciences : CABIOS
|April 1, 1992
PubMed
Summary

Machine-independent parallel programming offers portability benefits for biological sequence comparison tools with only a minor efficiency cost. This approach enhances software adaptability across diverse computing architectures.

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

Medical Informatics Training at Yale University School of Medicine.

Yearbook of medical informatics·2016
Same author

ATTENDING: Critiquing a Physician's Management Plan.

IEEE transactions on pattern analysis and machine intelligence·2011
Same author

The Common Data Elements for cancer research: remarks on functions and structure.

Methods of information in medicine·2006
Same author

Automating identification of adverse events related to abnormal lab results using standard vocabularies.

AMIA ... Annual Symposium proceedings. AMIA Symposium·2006
Same author

Metadata driven generation of reports for clinical studies.

AMIA ... Annual Symposium proceedings. AMIA Symposium·2006
Same author

Integration of Web-based and PC-based clinical research databases.

Methods of information in medicine·2004

Area of Science:

  • Computational Biology
  • Bioinformatics
  • Parallel Computing

Background:

  • Biological sequence comparison is crucial for understanding genetic relationships and protein functions.
  • Efficient parallel computation is essential for handling large-scale biological datasets.

Purpose of the Study:

  • To compare machine-independent versus machine-specific parallelization approaches for biological sequence comparison software.
  • To evaluate expressive power, portability, and efficiency of parallel programming strategies.

Main Methods:

  • Developed a platform program for biological sequence comparison with two routines: PSCANLIB and PCOMPLIB.
  • Parallelized the program using native Intel Hypercube commands (machine-specific).
  • Translated parallelization logic into Linda, a machine-independent parallel programming language.

Related Experiment Videos

Main Results:

  • The machine-independent approach (Linda) demonstrated superior portability across different parallel machines.
  • Both approaches were evaluated for expressive power in coordinating parallel computation.
  • Benchmark tests showed that machine-independent parallelization achieved benefits with only a modest reduction in efficiency.

Conclusions:

  • Machine-independent parallel programming, exemplified by Linda, provides significant advantages in software portability for bioinformatics tools.
  • The trade-off between portability and efficiency is acceptable, favoring the adaptable approach for broader applicability.
  • This study highlights the value of abstracting parallelization logic for scientific computing applications.