Dynamic Resource Scheduling And Optimization In Elastic Optical Networks

Author : Shuyi Yan
Publisher :
ISBN 13 :
Total Pages : 280 pages
Book Rating : 4.:/5 (972 download)

DOWNLOAD NOW!

Download or read book Analytical Frameworks to Evaluate Performance in Both Traditional and Elastic Optical Networks written by Shuyi Yan and published by . This book was released on 2016 with total page 280 pages. Available in PDF, EPUB and Kindle. Book excerpt: With the provisioning of network equipment becoming increasingly challenging due to the large number of application types that must be supported at once, it is generally expected that cloud based applications are going to experience improved end-to-end performance assurance while maintaining a competitive cost of service. Most applications must make use of various types of resources, including network bandwidth and server CPU cycles, among others. An application request may be blocked when one or more of the resource types that are required to achieve the desired QoS cannot be reserved due to their shortage. Elastic Optical Networks (EONs) enable optical circuits to be assigned distinct numbers of spectrum slices. Individual circuits can then be assigned an optimal number of slices to best match their target transmission rates. A well-known drawback of EONs is spectrum fragmentation and its resulting uneven blocking probability, which circuit requests experience when the available spectrum slices in the fiber are insufficient or not contiguous. Capturing this spectrum fragmentation problem analytically is a challenging problem. Not surprisingly, most of the existing studies at this time mainly use simulation-based techniques to quantify blocking probability in EONs. This dissertation proposes a Markov Chain (MC) model first to efficiently and accurately estimate the blocking probability caused by the multi-application and multi-resource (MA-MR) constraint. The strength of the model is its scalability and ability to account for hundreds of application types concurrently sharing multiple pools of distinct resources. Then a Steady State Probability (SSP) policy which is built upon this analytical MC model and a Network State (NS) policy are considered and proposed for joint allocation of network link and data center resources in the presence of dynamic application requests. The SSP policy is based on pre-computed probability to select data center and network resources and the NS policy is based on the knowledge of the network and data center state. An algorithm for optimally computing the probabilities based on the SSP policy is presented. Obtained simulation results demonstrate and numerically assess the trade-off between the system performance in terms of blocking probability and the signaling overhead. In order to quantify blocking probability theoretically in EONs, two Markov Chain (MC) models are presented to characterize the fragmentation problem in a simplified scenario, i.e., only two types of circuit services are allowed over a single fiber link, which is mentioned as two-service spectrum-elastic optical fiber link. One is an approximation model which is more effective and easier to be implemented based on the fully-shared fiber spectrum and then it is extended to account for partially-shared fiber spectrum. The other is an accurate model which is remarkably accurate to capture the fragmentation. The initial analytical efforts are able to capture the non-monotonic behavior of the blocking probability in EONs. Two product form based on analytical models are finally presented for quantifying blocking probability in a two-service elastic fiber link effectively. There is a trade-off between the accuracy and the model complexity.