DocumentCode :
3229930
Title :
An Innovative Restoration Algorithm with Prioritized Preemption Enabled
Author :
Wang, Jingyi ; Zhou, Zhiyu ; Wu, Jian
Author_Institution :
Beijing Univ. of Posts & Telecommun., Beijing
Volume :
3
fYear :
2007
fDate :
July 30 2007-Aug. 1 2007
Firstpage :
484
Lastpage :
489
Abstract :
This paper proposes an innovative prioritized restoration algorithm which enables carefully scheduled preemption (RAPPE). Firstly, it makes dedicate differentiation among network resources based on their roles in holding up services (working or backup) as well as the service levels they are carrying; secondly, it enables both higher level service preempting resources of lower and lower level service utilizing spare resources of higher; thirdly, it carefully schedules preempting rules. Simulation results show RAPPE satisfies various service requirements according to their priority levels and improves the network resource utilization sharply. Moreover, RAPPE also achieves wonderful performance in preemption perspective for it minimizes the preempted service level as well as the number of preempted services, both of which are important criteria in evaluating a preemption algorithm. Furthermore, to implement the RAPPE in existed networks, only a simple PathCost object needs to be installed on the existed routing module of edge nodes.
Keywords :
distributed processing; resource allocation; PathCost; RAPPE; innovative restoration algorithm; network resource utilization; prioritized preemption enabled; prioritized recovery model; Artificial intelligence; Costs; Delay; Distributed computing; Protection; Resource management; Routing; Scheduling algorithm; Software algorithms; Software engineering;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Software Engineering, Artificial Intelligence, Networking, and Parallel/Distributed Computing, 2007. SNPD 2007. Eighth ACIS International Conference on
Conference_Location :
Qingdao
Print_ISBN :
978-0-7695-2909-7
Type :
conf
DOI :
10.1109/SNPD.2007.462
Filename :
4287901
Link To Document :
بازگشت