| I. Matta and A. Udaya Shankar. Fast Time-Dependent Evaluation of Multi-Service Networks. Computer Networks and ISDN Systems - Special Issue on Modeling of Wired and Wireless ATM, Vol. 29, Nos. 17 & 18, February |
....policy, and the workload distribution. We compared several topology aggregation schemes that are di erent in terms of the accuracy of the routing information they provide. By means of the Z iteration method which numerically solves dynamic ow models of multi service connection oriented networks [11], we considered realistic skewed workloads as well as uniformly distributed workloads. Our simulation results led us to an interesting conclusion: under skewed workload, a less aggressive (and hence more costly) aggregation scheme, which results in a more detailed view of the network presented to ....
I. Matta and A. Udaya Shankar. Fast Time-Dependent Evaluation of Multi-Service Networks. Computer Networks and ISDN Systems - Special Issue on Modeling of Wired and Wireless ATM, Vol. 29, Nos. 17 & 18, February
....show how to apply it to PNNI. The effect of different aggregation schemes on network performance is compared by simulation in [3] However, they do not investigate the effect of workload distribution. Our Results We obtain our results using our recently developed time dependent evaluation method [16]. We consider both skewed and uniform workloads. In a skewed workload, some nodes are selected as the destination of the majority of the connections (or calls) and are thus considered hot spots. In a uniform workload, sourcedetination node pairs are uniformly distributed over the network. Under ....
....available bandwidth. Organization The rest of the paper is organized as follows. Section 2 presents our hierarchical network model. This model is solved using our recently developed timedependent evaluation method. The method solves dynamic flow models of multi service networks such as ATM see [16]. Section 3 discusses the computation and aggregation of routing metrics. Section 4 presents numerical results for a sample network. Section 5 concludes with future work. 2 Evaluation Model Network and Routing Models We consider a connection oriented network. To support real time QoS, necessary ....
I. Matta and A.U. Shankar. Fast Time-Dependent Evaluation of Multi-Service Networks. To appear in Computer Networks and ISDN Systems - Special Issue on Modeling of Wired and Wireless ATM. http://www.ccs.neu.edu/home/matta/publications.html.
....show how to apply it to PNNI. The effect of different aggregation schemes on network performance is compared by simulation in [3] However, the model only considers a single source destination pair. Our results: We obtain our results using our recently developed time dependent evaluation method [16]. We consider both skewed and uniform workloads. In a skewed workload, some nodes are selected as the destination of the majority of the connections (or calls) and are thus considered hot spots. In a uniform workload, source detination node pairs are uniformly distributed over the network. Under ....
....bandwidth. Organization: The rest of the paper is organized as follows. Section 2 presents our hierarchical network model. This model is solved using our recently developed time dependent evaluation method. The method solves dynamic flow models of multi service networks such as ATM see [16]. Section 3 discusses the computation and aggregation of routing metrics. Section 4 presents numerical results for a sample network, and concludes with future work. 2 Evaluation Model Network and routing models: We consider a connection oriented network. To support real time QoS, necessary ....
I. Matta and A. Udaya Shankar. Fast Time-Dependent Evaluation of Multi-Service Networks. To appear in Computer Networks and ISDN Systems - Special Issue on Modeling of Wired and Wireless ATM. http://www.ccs.neu.edu/home/matta/publications.html.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC