Search

We propose an MPLS switch with a GPS-based scheduler to provide guaranteed QoS in the Internet. The considered scenarios include cases with sufficient labels or with a limited number of labels under competing streams. Due to tractability issues in direct analysis, we use a worst case equivalent system model to obtain the theoretical upper bounds of delays and backlogs of the original system model. Simulation results are used to verify the analytical results. We can obtain the worst case delay bounds and the maximum backlogs. Furthermore, the trade-off among the number of labels, the worst delay bounds, and the maximum backlogs can be observed. With our model, one can determine the number of required labels according to the allowable delay and the available buffer size.

The performance of a GMPLS (generalized multiprotocol label switching) architecture with flush capability is studied. For this switching architecture, we propose a queueing model that includes the control plane, the switching buffer mechanism, and the flush mechanism. The flush capability is included to reduce the out-of-sequence problem due to dynamic path changes. The behavior of aggregated streams, the label-setup and release policies, and the mechanisms for efficient resource allocation are all covered. With the proposed model, one can select appropriate parameters for the label-setup policy and the label-release policy to match the traffic load and network environment. Key performance metrics, such as throughput, label-setup rate, and fast path bandwidth utilization, can all be evaluated by this mathematical model. Numerical results and simulations are used to verify the accuracy of our proposed queueing model. Finally, the trade-off among these performance metrics can be observed as well.

We propose a close-loop queueing model of a multiprotocol label switching (MPLS) switch with the different label-setup and release policies, supporting both traffic-driven and topology-driven connection setup procedures. This model can emulate the behavior of TCP under the MPLS switch when the maximum window size is sustained and the packet loss rate is negligible. From the proposed flow-based MPLS switch model, one can clearly observe the competition of multiple IP flow for limited labels, and how the label-setup policy and the label release policy affect the system performance. We find that Norton's theorem can be applied to solve this sophisticated queueing model. Therefore, with very limited computational complexity with respect to the number of IP flows or labels, the proposed mathematical model and the approximation of label competition can be used to obtain the desired performance metrics, such as the throughput, the label-setup rate, and the channel utilization. Finally, the trade-off among performance metrics can be observed as well.