This paper addresses the problem of optimal traffic engineering in a connectionless network. Based on Nonlinear control Theory, the approach taken in this paper provides a family of optimal adaptation laws. These laws...
详细信息
This paper addresses the problem of optimal traffic engineering in a connectionless network. Based on Nonlinear control Theory, the approach taken in this paper provides a family of optimal adaptation laws. These laws enable each node in the network to independently distribute traffic in an optimal way, as measured by a given utility function of a general form. Furthermore, this approach not only allows for multiple paths but also multiple Classes of Service. Moreover, the decentralized control scheme provided enables optimal traffic redistribution in the case of link failures. Suboptimal control laws are also presented in an effort to reduce the computational burden imposed on the nodes of the network.
We first study the impact of traffic burstiness on the accuracy of timed-stepped-simulation (TSS). Then we propose a compensation scheme to expand TSS's working ranges.
ISBN:
(纸本)0780378024
We first study the impact of traffic burstiness on the accuracy of timed-stepped-simulation (TSS). Then we propose a compensation scheme to expand TSS's working ranges.
Explicit multicasting (xcasting) is a type of multicasting that allows packets to be addressed to multiple destinations. Unfortunately, special routers and hosts for replicating and directing these packets are typical...
详细信息
Explicit multicasting (xcasting) is a type of multicasting that allows packets to be addressed to multiple destinations. Unfortunately, special routers and hosts for replicating and directing these packets are typically needed to deploy xcasting. One alternative approach to deployment is to use devices that can examine packets queued at routers and recode unicast traffic as xcast traffic. These devices allow xcasting to be added to the Internet incrementally and can optimize existing unicast traffic without requiring changes to existing hosts or applications. Simulations show that this type of xcasting performs well in reducing bandwidth usage and can even reduce network latency in certain situations. Although traditional multicasting offers superior scalability to xcasting, xcasting has many appealing properties worthy of further examination.
Link bundling is a way to increase routing scalability whenever a pair of label switching routers in MPLS are connected by multiple parallel links. However, link bundling can be inefficient as a label switched path (L...
详细信息
Link bundling is a way to increase routing scalability whenever a pair of label switching routers in MPLS are connected by multiple parallel links. However, link bundling can be inefficient as a label switched path (LSP) has to be associated with a particular link. In this paper, we show that the efficiency of link bundling can be significantly improved if traffic can be effectively distributed across the parallel links. We propose an ip switch architecture that is capable of distributing flows both inside the switch and among the parallel links based on operations that are relatively simple to implement. The switch requires no speedup, guarantees in-sequence packet delivery for a given flow, avoids complex coordination algorithms, and can achieve LSP throughput higher than the line rate. By means of simulation using ip traces, we investigate the performance of the proposed switch, and show that the switch achieves good load-balancing performance. We describe extensions to the basic architecture which allows for very large bundle size, handles incremental upgrade strategy, improves reliability, and accommodates noniptraffic.
In the current development of all ipnetworks different traffic classes, which are used for different services will share the same bandwidth resources of the network. For effective usage of the system resources, arriv...
详细信息
In the current development of all ipnetworks different traffic classes, which are used for different services will share the same bandwidth resources of the network. For effective usage of the system resources, arrivals for some low priority traffic classes could be rejected in cases when risk of having overload is high. In this paper, the use of connection admission control (CAC) algorithm together with scheduler optimizing is considered. The scheduler is functioning according to the optimal weight allocation for each traffic class, and CAC is based on the current utilization of the scheduler. Also one traffic class, which is referenced as "voice" class has more priority compared to others. For the scheduler optimization, revenue function is considered with linear approximation and the adaptive weights are found based on Lagrangian relaxation. To show the accuracy of the use of CAC algorithm, simulation results are presented and analyzed.
Sip is evolving as the dominant protocol for multimedia call control in ipnetworks and is expected to be widely deployed in the near future. Using Sip for supporting mobility in Sip-based networks appears as a very a...
详细信息
Sip is evolving as the dominant protocol for multimedia call control in ipnetworks and is expected to be widely deployed in the near future. Using Sip for supporting mobility in Sip-based networks appears as a very attractive alternative to mobile ip, taking advantage of existing Sip infrastructure and signaling, while avoiding duplication of functionality. However, existing proposals for supporting mobility with Sip are not efficient in handling micro-mobility and do not cater for all types of traffic. In this paper, we introduce hierarchical mobile Sip (HMSip), for efficient micro-mobility management in Sip environments. HMSip is a Sip-based scheme that builds on existing ip mobility protocols, aiming at integrating their key concepts in the most beneficial way. Our proposed scheme minimizes the handoff latency and the backbone signaling overhead, while catering for all types of traffic. Moreover, HMSip can be effectively combined with micro-mobility schemes for QoS flows, producing, thus, a complete micro-mobility solution for Sip flows with QoS guarantees.
MPLS has been enhanced with the capability to precisely engineer traffic tunnels to avoid congestion and utilize all available bandwidth in an efficient manner. The distributed fashion of setting up the labels and for...
详细信息
MPLS has been enhanced with the capability to precisely engineer traffic tunnels to avoid congestion and utilize all available bandwidth in an efficient manner. The distributed fashion of setting up the labels and formation of data tunnels for the assigned FEC (forwarding equivalence class) is a highly sophisticated mechanism, involving great dexterity on the part of label switched routers. This paper provides an extensive resource, serving the motivation for loop detection in MPLS. This proposal is in conjunction with the colored thread algorithm (Ohba et al. (2001)), and addresses the issues of efficiency in both time and memory usage in setting up the label switched paths (LSP). The approach taken is to use the global variables, instead of ip addresses. The scheme is a loop prevention scheme, thus ensuring loop detection and loop mitigation. Furthermore, a method of assigning the labels, while rewinding the thread has been explained, which could substantially reduce the LSP set up time and add to the efficient thread mechanism.
In this paper, we propose, develop and evaluate a new adaptive sampling scheme for monitoring and measuring network performance metrics in MPLS-based ipnetworks supporting real time applications. The paper is focused...
详细信息
In this paper, we propose, develop and evaluate a new adaptive sampling scheme for monitoring and measuring network performance metrics in MPLS-based ipnetworks supporting real time applications. The paper is focused on investigating the mechanisms that can adaptively adjust the parameters of the sampling technique based on the estimated traffic rate. The performance of the proposed technique is compared with conventional sampling method by conducting simulation experiments using voice traffic patterns. Simulation results are presented to illustrate that adaptive sampling provides the potential for better monitoring, control, and management of high-performance networks with higher accuracy, lower overhead, or both.
In this paper, we introduce and analyze a feedback control model of TCP/AQM dynamics. Then, we propose the pro-active queue management (PAQM) mechanism, which can provide proactive congestion avoidance and control usi...
详细信息
ISBN:
(纸本)0780377885
In this paper, we introduce and analyze a feedback control model of TCP/AQM dynamics. Then, we propose the pro-active queue management (PAQM) mechanism, which can provide proactive congestion avoidance and control using an adaptive congestion indicator and a control function for wide range of traffic environments. The PAQM stabilizes the queue length around a desired level while giving smooth and low packet loss rates independent of the traffic load level under a wide range of traffic environments. The PAQM outperforms other AQM algorithms such as random early detection (RED) [S. Floyd and V. Jacobson, August 1993] and PI-controller [C. V. Hollot et al., April 2001].
In this paper, a congestion control algorithm for best-effort traffic in ipnetworks is proposed. The aim is to avoid congestion and loss of packets, and also to share the bandwidth of a congested link among the flows...
详细信息
In this paper, a congestion control algorithm for best-effort traffic in ipnetworks is proposed. The aim is to avoid congestion and loss of packets, and also to share the bandwidth of a congested link among the flows that are using it. Each client provides the flow rate value that would like to obtain, and a time value (round-trip delay) that is used by routers to compute the number of active flows. One assumes that routers are allowed to use a field in the packets to propagate the reference of the flow rate, in principle, it can be done using the field option of the ip header. The clients use the information that is provided by the network and the information that is obtained by probing the Internet to adjust the servers' flow rates. In the paper on follows the principle that clients do not need to increase their flow rates to obtain information about the congestion state of the network.
暂无评论