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

Long-term forecasting of internet backbone traffic.

Konstantina Papagiannaki1, Nina Taft, Zhi-Li Zhang

  • 1Sprint ATL, Intel Research, Cambridge CB3 OFD, UK. dina.papagiannaki@intel.com

IEEE Transactions on Neural Networks
|October 29, 2005
PubMed
Summary

This study presents a method to predict Internet backbone network upgrades using SNMP data. Wavelet analysis and time series models accurately forecast traffic demand for capacity planning up to six months ahead.

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

Chlorogenic Acid Alleviates Experimental Asthma by Reprogramming DHA Metabolism to Inhibit Ferroptosis.

International journal of molecular sciences·2026
Same author

High-resolution computed tomography predicts optimal cochlear implantation strategy in patients with chronic otitis media.

World journal of radiology·2026
Same author

High-Efficiency Gold Nanoaggregates for NIR LED-Driven Sustained Mild Photothermal Therapy Achieving Complete Tumor Eradication and Immune Enhancement.

Advanced materials (Deerfield Beach, Fla.)·2024
Same author

Development and validation of a risk prediction model for community-acquired pressure injury in a cancer population: A case-control study.

Journal of tissue viability·2024
Same author

Efficacy of high-voltage, long-duration pulsed radiofrequency for the treatment of acute herpes zoster-related trigeminal neuralgia in a patient with permanent pacemaker.

The Korean journal of pain·2024
Same author

A mitochondria-targeting dihydroartemisinin derivative as a reactive oxygen species -based immunogenic cell death inducer.

iScience·2024

Area of Science:

  • Network Engineering
  • Data Science
  • Telecommunications

Background:

  • Internet backbone networks require proactive capacity planning to manage increasing traffic demands.
  • Accurate prediction of link additions and upgrades is crucial for efficient network operation and cost management.

Purpose of the Study:

  • To develop a methodology for predicting the timing and location of necessary link additions/upgrades in IP backbone networks.
  • To leverage historical network traffic data for long-term demand forecasting.

Main Methods:

  • Utilized Simple Network Management Protocol (SNMP) statistics collected since 1999.
  • Applied wavelet multiresolution analysis (MRA) to identify long-term trends and traffic variability at multiple time scales.
  • Employed linear time series models, including ARIMA, to forecast traffic demand components.

Related Experiment Videos

Main Results:

  • IP backbone traffic exhibits clear long-term trends, strong periodicities, and multi-scale variability, with 12-hour fluctuations being significant.
  • A regression model incorporating trend and 12-hour fluctuations explained 90% of traffic variance.
  • Forecasts using ARIMA models accurately predicted traffic trends and fluctuations for at least six months.

Conclusions:

  • The proposed methodology provides accurate and reliable long-term traffic demand forecasts for IP backbone networks.
  • This enables proactive planning for link upgrades, optimizing network performance and resource allocation.
  • The approach effectively combines wavelet analysis and time series modeling for network capacity management.