東華大學圖書館 |

Resource Allocation Algorithms for Elastic Optical Networks.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Resource Allocation Algorithms for Elastic Optical Networks./
作者:	Afsharlar, Pegah.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2021,
面頁冊數:	113 p.
附註:	Source: Dissertations Abstracts International, Volume: 82-08, Section: B.
Contained By:	Dissertations Abstracts International82-08B.
標題:	Communication. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28260796
ISBN:	9798569981311

Resource Allocation Algorithms for Elastic Optical Networks.
Afsharlar, Pegah.

Resource Allocation Algorithms for Elastic Optical Networks. - Ann Arbor : ProQuest Dissertations & Theses, 2021 - 113 p.

Source: Dissertations Abstracts International, Volume: 82-08, Section: B.

Thesis (Ph.D.)--University of Massachusetts Lowell, 2021.

This item must not be sold to any third party vendors.

Due to unprecedented rapid growth of data centers, ubiquitous cloud computing, and the necessity to use high-speed networks for transferring large datasets (approaching petabyte and exabyte scales), traditional rigid Wavelength routed networks are proving insufficient for emerging applications. Wavelength routed networks are limited by the coarse granularity at which spectral resources may be consumed. A promising solution for next-generation high-speed optical transport that provides greater levels of flexibility and efficiency to the spectral domain is the Elastic Optical Network (EON). EONs treat the available optical spectrum as an elastic resource and permits allocation of portions of the spectrum range in a variable and adaptive manner as needed for efficient consumption. To do so, traffic demands need to be routed over the network and a portion of the optical spectrum allocated to each of them thus, creating lightpaths. Routing and spectrum allocation (RSA) problem is an important optimization problem that needs to be solved. Network operators can reserve spectral resources for applications in advance of service as an alternative to demanding solutions without scheduling foresight. These types of requests are called Advance Reservation (AR) requests. AR has several important applications for wide-area networks. It can be used to provide high-bandwidth services such as video conferencing and in applications requiring the scheduled distribution of large files. AR can also be beneficial to the network by allowing the network operator to better plan resource usage across a varied array of traffic inputs and thereby increase overall utilization success. Knowledge of specific reservation service-times can lead to enhanced optimization opportunities such as resource utilization and blocking probability. Many applications involve delay-tolerant background or recurring tasks. By providing advance reservation for such tasks, the network can achieve higher utilizationwhile increasing the probability that the applications will be able to successfully reserve the required network resources. The major objective of this thesis is to study the problem of RSA in EONs and improve the resource scheduling efficiency among a set of competing and overlapping demands on a core network with realistically limited spectral resources. A novel RSA technique for AR traffic called Delayed Spectrum Allocation (DSA) is proposed. The hypothesis is that by delaying the allocation of spectral resources until just before the actual transmission begins, though can more spectrum efficiently perform scheduling for a set of temporally overlapping demands. Several DSA approaches for dynamically arriving AR traffic is qualitatively and quantitatively evaluated through extensive simulation and modeling techniques. DSA is compared with traditional and new rescheduling alternatives. DSA exhibits the tremendous potential to reduce blocking for scheduled demands.Another important contribution of this dissertation is proposing, the novel integration of both the delay allocation and lightpath switching techniques in order to investigate the impact of supporting both enhancements in tandem. This technic is evaluated for WDM network and EONs. A three-phase resource provisioning and allocation algorithm called Delayed Spectrum Allocation with Spectrum Switching (DSA-SS), is developed and evaluated both qualitatively and quantitatively against baseline solutions supporting only DSA. The novel approach of combining these enhancements significantly increase the performance improvement of DSA.

ISBN: 9798569981311Subjects--Topical Terms:

524709
Communication.
Subjects--Index Terms:

Advanced reservation

Resource Allocation Algorithms for Elastic Optical Networks.
LDR:04878nmm a2200409 4500 001 2343141
005 20220502104144.5
008 241004s2021 ||||||||||||||||| ||eng d
020 $a 9798569981311
035 $a (MiAaPQ)AAI28260796
035 $a AAI28260796
040 $a MiAaPQ $c MiAaPQ
100 1 $a Afsharlar, Pegah. $3 3681600
245 1 0 $a Resource Allocation Algorithms for Elastic Optical Networks.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2021
300 $a 113 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-08, Section: B.
500 $a Advisor: Vokkarane, Vinod;Armiento, Craig.
502 $a Thesis (Ph.D.)--University of Massachusetts Lowell, 2021.
506 $a This item must not be sold to any third party vendors.
520 $a Due to unprecedented rapid growth of data centers, ubiquitous cloud computing, and the necessity to use high-speed networks for transferring large datasets (approaching petabyte and exabyte scales), traditional rigid Wavelength routed networks are proving insufficient for emerging applications. Wavelength routed networks are limited by the coarse granularity at which spectral resources may be consumed. A promising solution for next-generation high-speed optical transport that provides greater levels of flexibility and efficiency to the spectral domain is the Elastic Optical Network (EON). EONs treat the available optical spectrum as an elastic resource and permits allocation of portions of the spectrum range in a variable and adaptive manner as needed for efficient consumption. To do so, traffic demands need to be routed over the network and a portion of the optical spectrum allocated to each of them thus, creating lightpaths. Routing and spectrum allocation (RSA) problem is an important optimization problem that needs to be solved. Network operators can reserve spectral resources for applications in advance of service as an alternative to demanding solutions without scheduling foresight. These types of requests are called Advance Reservation (AR) requests. AR has several important applications for wide-area networks. It can be used to provide high-bandwidth services such as video conferencing and in applications requiring the scheduled distribution of large files. AR can also be beneficial to the network by allowing the network operator to better plan resource usage across a varied array of traffic inputs and thereby increase overall utilization success. Knowledge of specific reservation service-times can lead to enhanced optimization opportunities such as resource utilization and blocking probability. Many applications involve delay-tolerant background or recurring tasks. By providing advance reservation for such tasks, the network can achieve higher utilizationwhile increasing the probability that the applications will be able to successfully reserve the required network resources. The major objective of this thesis is to study the problem of RSA in EONs and improve the resource scheduling efficiency among a set of competing and overlapping demands on a core network with realistically limited spectral resources. A novel RSA technique for AR traffic called Delayed Spectrum Allocation (DSA) is proposed. The hypothesis is that by delaying the allocation of spectral resources until just before the actual transmission begins, though can more spectrum efficiently perform scheduling for a set of temporally overlapping demands. Several DSA approaches for dynamically arriving AR traffic is qualitatively and quantitatively evaluated through extensive simulation and modeling techniques. DSA is compared with traditional and new rescheduling alternatives. DSA exhibits the tremendous potential to reduce blocking for scheduled demands.Another important contribution of this dissertation is proposing, the novel integration of both the delay allocation and lightpath switching techniques in order to investigate the impact of supporting both enhancements in tandem. This technic is evaluated for WDM network and EONs. A three-phase resource provisioning and allocation algorithm called Delayed Spectrum Allocation with Spectrum Switching (DSA-SS), is developed and evaluated both qualitatively and quantitatively against baseline solutions supporting only DSA. The novel approach of combining these enhancements significantly increase the performance improvement of DSA.
590 $a School code: 0111.
650 4 $a Communication. $3 524709
650 4 $a Computer engineering. $3 621879
650 4 $a Electrical engineering. $3 649834
650 4 $a Optics. $3 517925
653 $a Advanced reservation
653 $a Anycast
653 $a Delayed allocation
653 $a Elastic optical network
653 $a EON
653 $a Lightpath switching
653 $a Wavelength routed networks
690 $a 0459
690 $a 0464
690 $a 0752
690 $a 0544
710 2 $a University of Massachusetts Lowell. $b Computer Engineering. $3 3553410
773 0 $t Dissertations Abstracts International $g 82-08B.
790 $a 0111
791 $a Ph.D.
792 $a 2021
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28260796