Realistic data transfer scheduling with uncertainty

Upendra Rathnayake, Mohsin Iftikhar, Maximilian Ott, Aruna Prasad Seneviratne

Research output: Contribution to journalArticle

2 Citations (Scopus)
5 Downloads (Pure)

Abstract

Next Generation Networks (NGNs) will be comprised of different access technologies. We are already seeing the emergence of mobile devices with the capability of connecting to heterogeneous networks with different capabilities and constraints. In addition, many bandwidth intensive applications have rather relaxed real-time constraints allowing for alternative scheduling mechanisms which can take into account user preferences, network characteristics as well as future network resource availability to better exploit network heterogeneity. The current approaches either simply react to changes, or assume that availability predictions are perfect. In this paper, we propose a scheduling scheme based on stochastic modeling to account for prediction errors. The scheme optimizes overall user utility gain considering imperfect predictions taken over realistic time intervals while catering for different applications' needs. We use 180 days of real user data of many users to demonstrate that it consistently outperforms other non-stochastic and greedy approaches in typical networking environments.
Original languageEnglish
Pages (from-to)1055-1065
Number of pages11
JournalComputer Communications
Volume34
Issue number9
DOIs
Publication statusPublished - 2011

Fingerprint

Data transfer
Scheduling
Availability
Next generation networks
Heterogeneous networks
Mobile devices
Bandwidth
Uncertainty

Cite this

Rathnayake, Upendra ; Iftikhar, Mohsin ; Ott, Maximilian ; Seneviratne, Aruna Prasad. / Realistic data transfer scheduling with uncertainty. In: Computer Communications. 2011 ; Vol. 34, No. 9. pp. 1055-1065.
@article{fa8e405094ce4a5398073f90861d47a7,
title = "Realistic data transfer scheduling with uncertainty",
abstract = "Next Generation Networks (NGNs) will be comprised of different access technologies. We are already seeing the emergence of mobile devices with the capability of connecting to heterogeneous networks with different capabilities and constraints. In addition, many bandwidth intensive applications have rather relaxed real-time constraints allowing for alternative scheduling mechanisms which can take into account user preferences, network characteristics as well as future network resource availability to better exploit network heterogeneity. The current approaches either simply react to changes, or assume that availability predictions are perfect. In this paper, we propose a scheduling scheme based on stochastic modeling to account for prediction errors. The scheme optimizes overall user utility gain considering imperfect predictions taken over realistic time intervals while catering for different applications' needs. We use 180 days of real user data of many users to demonstrate that it consistently outperforms other non-stochastic and greedy approaches in typical networking environments.",
keywords = "Heterogeneous networking with mobility, Optimization with multi-interfaces, Two-stage stochastic linear program",
author = "Upendra Rathnayake and Mohsin Iftikhar and Maximilian Ott and Seneviratne, {Aruna Prasad}",
note = "Imported on 12 Apr 2017 - DigiTool details were: Journal title (773t) = Computer Communications. ISSNs: 0140-3664;",
year = "2011",
doi = "10.1016/j.comcom.2010.02.012Document",
language = "English",
volume = "34",
pages = "1055--1065",
journal = "Computer Communications",
issn = "0140-3664",
publisher = "Elsevier",
number = "9",

}

Realistic data transfer scheduling with uncertainty. / Rathnayake, Upendra; Iftikhar, Mohsin; Ott, Maximilian; Seneviratne, Aruna Prasad.

In: Computer Communications, Vol. 34, No. 9, 2011, p. 1055-1065.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Realistic data transfer scheduling with uncertainty

AU - Rathnayake, Upendra

AU - Iftikhar, Mohsin

AU - Ott, Maximilian

AU - Seneviratne, Aruna Prasad

N1 - Imported on 12 Apr 2017 - DigiTool details were: Journal title (773t) = Computer Communications. ISSNs: 0140-3664;

PY - 2011

Y1 - 2011

N2 - Next Generation Networks (NGNs) will be comprised of different access technologies. We are already seeing the emergence of mobile devices with the capability of connecting to heterogeneous networks with different capabilities and constraints. In addition, many bandwidth intensive applications have rather relaxed real-time constraints allowing for alternative scheduling mechanisms which can take into account user preferences, network characteristics as well as future network resource availability to better exploit network heterogeneity. The current approaches either simply react to changes, or assume that availability predictions are perfect. In this paper, we propose a scheduling scheme based on stochastic modeling to account for prediction errors. The scheme optimizes overall user utility gain considering imperfect predictions taken over realistic time intervals while catering for different applications' needs. We use 180 days of real user data of many users to demonstrate that it consistently outperforms other non-stochastic and greedy approaches in typical networking environments.

AB - Next Generation Networks (NGNs) will be comprised of different access technologies. We are already seeing the emergence of mobile devices with the capability of connecting to heterogeneous networks with different capabilities and constraints. In addition, many bandwidth intensive applications have rather relaxed real-time constraints allowing for alternative scheduling mechanisms which can take into account user preferences, network characteristics as well as future network resource availability to better exploit network heterogeneity. The current approaches either simply react to changes, or assume that availability predictions are perfect. In this paper, we propose a scheduling scheme based on stochastic modeling to account for prediction errors. The scheme optimizes overall user utility gain considering imperfect predictions taken over realistic time intervals while catering for different applications' needs. We use 180 days of real user data of many users to demonstrate that it consistently outperforms other non-stochastic and greedy approaches in typical networking environments.

KW - Heterogeneous networking with mobility

KW - Optimization with multi-interfaces

KW - Two-stage stochastic linear program

U2 - 10.1016/j.comcom.2010.02.012Document

DO - 10.1016/j.comcom.2010.02.012Document

M3 - Article

VL - 34

SP - 1055

EP - 1065

JO - Computer Communications

JF - Computer Communications

SN - 0140-3664

IS - 9

ER -