skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Time-Series Forecast Modeling on High-Bandwidth Network Measurements

Abstract

With the increasing number of geographically distributed scientific collaborations and the growing sizes of scientific data, it has become challenging for users to achieve the best possible network performance on a shared network. In this paper, we have developed a model to forecast expected bandwidth utilization on high-bandwidth wide area networks. The forecast model can improve the efficiency of the resource utilization and scheduling of data movements on high-bandwidth networks to accommodate ever increasing data volume for large-scale scientific data applications. A univariate time-series forecast model is developed with the Seasonal decomposition of Time series by Loess (STL) and the AutoRegressive Integrated Moving Average (ARIMA) on Simple Network Management Protocol (SNMP) path utilization measurement data. Compared with the traditional approach such as Box-Jenkins methodology to train the ARIMA model, our forecast model reduces computation time up to 92.6 %. It also shows resilience against abrupt network usage changes. Finally, our forecast model conducts the large number of multi-step forecast, and the forecast errors are within the mean absolute deviation (MAD) of the monitored measurements.

Authors:
 [1];  [1]
  1. Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Publication Date:
Research Org.:
Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
OSTI Identifier:
1377437
Grant/Contract Number:
AC02-05CH11231
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Journal of Grid Computing
Additional Journal Information:
Journal Volume: 14; Journal Issue: 3; Journal ID: ISSN 1570-7873
Publisher:
Springer
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; data modeling; time series; prediction model; network measurements; network traffic

Citation Formats

Yoo, Wucherl, and Sim, Alex. Time-Series Forecast Modeling on High-Bandwidth Network Measurements. United States: N. p., 2016. Web. doi:10.1007/s10723-016-9368-9.
Yoo, Wucherl, & Sim, Alex. Time-Series Forecast Modeling on High-Bandwidth Network Measurements. United States. doi:10.1007/s10723-016-9368-9.
Yoo, Wucherl, and Sim, Alex. Fri . "Time-Series Forecast Modeling on High-Bandwidth Network Measurements". United States. doi:10.1007/s10723-016-9368-9. https://www.osti.gov/servlets/purl/1377437.
@article{osti_1377437,
title = {Time-Series Forecast Modeling on High-Bandwidth Network Measurements},
author = {Yoo, Wucherl and Sim, Alex},
abstractNote = {With the increasing number of geographically distributed scientific collaborations and the growing sizes of scientific data, it has become challenging for users to achieve the best possible network performance on a shared network. In this paper, we have developed a model to forecast expected bandwidth utilization on high-bandwidth wide area networks. The forecast model can improve the efficiency of the resource utilization and scheduling of data movements on high-bandwidth networks to accommodate ever increasing data volume for large-scale scientific data applications. A univariate time-series forecast model is developed with the Seasonal decomposition of Time series by Loess (STL) and the AutoRegressive Integrated Moving Average (ARIMA) on Simple Network Management Protocol (SNMP) path utilization measurement data. Compared with the traditional approach such as Box-Jenkins methodology to train the ARIMA model, our forecast model reduces computation time up to 92.6 %. It also shows resilience against abrupt network usage changes. Finally, our forecast model conducts the large number of multi-step forecast, and the forecast errors are within the mean absolute deviation (MAD) of the monitored measurements.},
doi = {10.1007/s10723-016-9368-9},
journal = {Journal of Grid Computing},
number = 3,
volume = 14,
place = {United States},
year = {Fri Jun 24 00:00:00 EDT 2016},
month = {Fri Jun 24 00:00:00 EDT 2016}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record

Citation Metrics:
Cited by: 1work
Citation information provided by
Web of Science

Save / Share:
  • With the increasing number of geographically distributed scientific collaborations and the scale of the data size growth, it has become more challenging for users to achieve the best possible network performance on a shared network. We have developed a forecast model to predict expected bandwidth utilization for high-bandwidth wide area network. The forecast model can improve the efficiency of resource utilization and scheduling data movements on high-bandwidth network to accommodate ever increasing data volume for large-scale scientific data applications. Univariate model is developed with STL and ARIMA on SNMP path utilization data. Compared with traditional approach such as Box-Jenkins methodology,more » our forecast model reduces computation time by 83.2percent. It also shows resilience against abrupt network usage change. The accuracy of the forecast model is within the standard deviation of the monitored measurements.« less
  • With the increasing number of geographically distributed scientific collaborations and the scale of the data size growth, it has become more challenging for users to achieve the best possible network performance on a shared network. We have developed a forecast model to predict expected bandwidth utilization for high-bandwidth wide area network. The forecast model can improve the efficiency of resource utilization and scheduling data movements on high-bandwidth network to accommodate ever increasing data volume for large-scale scientific data applications. Univariate model is developed with STL and ARIMA on SNMP path utilization data. Compared with traditional approach such as Box-Jenkins methodology,more » our forecast model reduces computation time by 83.2%. It also shows resilience against abrupt network usage change. The accuracy of the forecast model is within the standard deviation of the monitored measurements.« less
  • Generation and prediction of time series are analyzed for the case of a bit generator: a perceptron where in each time step the input units are shifted one bit to the right with the state of the leftmost input unit set equal to the output unit in the previous time step. The long-time dynamical behavior of the bit generator consists of cycles whose typical period scales polynomially with the size of the network and whose spatial structure is periodic with a typical finite wavelength. The generalization error on a cycle is zero for a finite training set, and global dynamicalmore » behaviors can also be learned in a finite time. Hence, a projection of a rule can be learned in a finite time.« less
  • In the present study, we applied the methodology of the complex network-based time series analysis to experimental temperature time series from a vertical turbulent heated jet. More specifically, we approach the hydrodynamic problem of discriminating time series corresponding to various regions relative to the jet axis, i.e., time series corresponding to regions that are close to the jet axis from time series originating at regions with a different dynamical regime based on the constructed network properties. Applying the transformation phase space method (k nearest neighbors) and also the visibility algorithm, we transformed time series into networks and evaluated the topologicalmore » properties of the networks such as degree distribution, average path length, diameter, modularity, and clustering coefficient. The results show that the complex network approach allows distinguishing, identifying, and exploring in detail various dynamical regions of the jet flow, and associate it to the corresponding physical behavior. In addition, in order to reject the hypothesis that the studied networks originate from a stochastic process, we generated random network and we compared their statistical properties with that originating from the experimental data. As far as the efficiency of the two methods for network construction is concerned, we conclude that both methodologies lead to network properties that present almost the same qualitative behavior and allow us to reveal the underlying system dynamics.« less