Browsing by Subject "Queue"

Now showing 1 - 2 of 2

Open Access
AIMD-based online MPLS traffic engineering for TCP flows via distributed multi-path routing
(Springer, 2004) Alparslan O.; Akar, N.; Karasan, E.
With this paper, we propose a distributed online traffic engineering architecture for MPLS networks. In this architecture, a primary and secondary MPLS LSP are established from an ingress LSR to every other egress LSR. We propose to split the TCP traffic between the primary and secondary paths using a distributed mechanism based on ECN marking and AIMD-based rate control. Inspired by the random early detection mechanism for active queue management, we propose a random early reroute scheme to adaptively control the delay difference between the primary and secondary LSPS. Considering the adverse effect of packet reordering on TCP performance for packet-based load balancing schemes, we propose that the TCP splitting mechanism operates on a per-flow basis. Using flow-based models developed for Internet traffic and simulations, we show that flow-based distributed multi-path traffic engineering outperforms on a consistent basis the case of a single path in terms of per-flow goodputs. Due to the elimination of out-of-order packet arrivals, flow-based splitting also enhances TCP performance with respect to packet-based splitting especially for long TCP flows that are hit hard by packet reordering. We also compare and contrast two queuing architectures for differential treatment of data packets routed over primary and secondary LSPS in the MPLS data plane, namely first-in-first-out and strict priority queuing. We show through simulations that strict priority queuing is more effective and relatively more robust with respect to the changes in the traffic demand matrix than first-in-first-out queuing in the context of distributed multi-path routing.
Open Access
Coordination of staffing and pricing decisions in a service firm
(John Wiley & Sons, 2008) Serel, D. A.; Erel, E.
Customer demand is sensitive to the price paid for the service in many service environments. Using queueing theory framework, we develop profit maximization models for jointly determining the price and the staffing level in a service company. The models include constraints on the average waiting time and the blocking probability. We show convexity of the single-variable subproblem under certain plausible assumptions on the demand and staffing cost functions. Using numerical examples, we investigate the sensitivity of the price and the staffing level to changes in the marginal service cost and the user-specified constraint on the congestion measure.