A New Approach to Routing with Dynamic Metrics

2006

This paper considers the use of compound cost functions in routing calculations. Using an abstracted version of Cisco’s EIGRP as its basic model, it develops the theoretical principals of optimal end-to-end interior routing then details the limitations of conventional and current implementation. The requirements of an improved system are discussed and proposals for an enhanced Ant Colony Optimisation - DUAL

1988

Attempting to dynamically adapt network parameters to the load seen can produce unexpected results. We present a simple model network example which demonstrates unstable behavior when traffic is directed according to routing optimized for minimal delay and the load varies at a rate comparable to the routing calculation time. The instability can be avoided by using almost any alternate design which avoids the knees on the delay curves, e.g. a Maximum Entropy Method design. The delay penalty in this case turns out to be small. This paper is very mathematical, but the point is simple: attempting make network parameters change as quickly as possible may not be an appropriate network management strategy.

2001

We propose a dynamic multipath routing (DMPR) scheme to improve resource utilization of a network for a given quality of service (QoS) in carrying delay sensitive traffic. The proposed scheme takes advantage of available alternate routes which can be used beneficially in communication networks or switches, without increasing the network latency. Through simulation study of circuit-switched, virtual circuitswitched (ATM) and packet-switched (Internet) networks we demonstrate the benefit of using DMPR over single-path routing (SPR) scheme. We also discuss the implementation aspects of DMPR scheme in data networks and the overhead associated with it.

Computer Communication Review, 1999

Internet service providers face a daunting challenge in provisioning network resources, due to the rapid growth of the Internet and wide fluctuations in the underlying traffic patterns. The ability of dynamic routing to circumvent congested links and improve application performance makes it a valuable traffic engineering tool. However, deployment of load-sensitive routing is hampered by the overheads imposed by link-state update propagation, path selection, and signalling. Under reasonable protocol and computational overheads, traditional approaches to load-sensitive routing of IP traffic are ineffective, and can introduce significant route flapping, since paths are selected based on out-of-date link-state information. Although stability is improved by performing load-sensitive routing at the flow level, flapping still occurs, because most IP flows have a short duration relative to the desired frequency of link-state updates. To address the efficiency and stability challenges of load-sensitive routing, we introduce a new hybrid approach that performs dynamic routing of long-lived flows, while forwarding short-lived flows on static preprovisioned paths. By relating the detection of long-lived flows to the timescale of link-state update messages in the routing protocol, route stability is considerably improved. Through simulation experiments using a one-week ISP packet trace, we show that our hybrid approach significantly outperforms traditional static and dynamic routing schemes, by reacting to fluctuations in network load without introducing route flapping.

ACM SIGSIM Simulation Digest, 1992

This paper discusses a new distributed, dynamic routing approach for communication networks. At each node, the best available route to the destination is selected and the message is forwarded to the next node on that route. The best route is evaluated at the new node again. The choice of the best route is based on a parameter K, defined in the paper. This parameter is calculated for every path based on the characteristics of the path, such as-the number of hops, the speeds of the links, the load on the links, reliability of links etc. The method guarantees that as long as the network is connected the message will be delivered to the destination using the best available path. This approach does not have the counting to infinity problem and is loop-free. A comparative study of this scheme with some of the existing schemes is also included to demonstrate its generality and effectiveness. I. INTRODUCTION

2010 IEEE Network Operations and Management Symposium - NOMS 2010, 2010

Intradomain routing in IP networks follows shortest paths according to administrative link costs. When several equalcost shortest paths exist, routers that use equal-cost multipath (ECMP) distribute the traffic over all of them. To produce singleshortest path (SSP) routing, a selection mechanism (tie-breaker) chooses just one of the equal-cost paths. Tie-breakers are poorly standardized and use information that may change over time, which makes SSP routing unpredictable. Therefore, link costs producing unique shortest paths (USP) are preferred.

2008

Despite the bursty and highly volatile traffic, routing in the Internet today is optimised only on coarse time scales, as load-adaptive routing is known to induce performance deterioration by causing massive oscillations. We describe ReplEx, an universally applicable distributed algorithm for dynamic routing/traffic engineering, which is based on game theory. We show through extensive realistic simulations that ReplEx does not oscillate, and that it achieves performance gains comparable to traditional static traffic engineering. To obtain a better understanding of the traffic to be routed, we furthermore analyse global traffic matrices and search engine induced traffic, based on traffic measurements. Moreover we describe EaC, a universally applicable memory efficient algorithm that identify the most popular nodes in a search tree, which can serve to pin down main traffic contributors.

IEEE Globecom 2006, 2006

We consider the problem of finding a single-path intra-domain routing for time-varying traffic. We characterize the traffic variations by a finite set of traffic profiles with given non-zero fractions of occurrence. Our goal is to optimize the average performance over all of these traffic profiles. We solve the optimal multi-path version of this problem using linear programming and develop heuristic single-path solutions using randomized rounding and iterated rounding.

Physical Review E, 2006

The efficiency of traffic routing on complex networks can be reflected by two key measurements, i.e., the network capacity and the average travel time of data packets. In this paper we propose a mixing routing strategy by integrating local static and dynamic information for enhancing the efficiency of traffic on scale-free networks. The strategy is governed by a single parameter. Simulation results show that maximizing the network capacity and reducing the packet travel time can generate an optimal parameter value. Compared with the strategy of adopting exclusive local static information, the new strategy shows its advantages in improving the efficiency of the system. The detailed analysis of the mixing strategy is provided for explaining its effects on traffic routing. The work indicates that effectively utilizing the larger degree nodes plays a key role in scalefree traffic systems.

ACM SIGCOMM Computer Communication Review, 2006

Extensive measurement studies have shown that end-to-end Internet path performance degradation is correlated with routing dynamics. However, the root cause of the correlation between routing dynamics and such performance degradation is poorly understood. In particular, how do routing changes result in degraded end-to-end path performance in the first place? How do factors such as topological properties, routing policies, and iBGP configurations affect the extent to which such routing events can cause performance degradation? Answers to these questions are critical for improving network performance.In this paper, we conduct extensive measurement that involves both controlled routing updates through two tier-1 ISPs and active probes of a diverse set of end-to-end paths on the Internet. We find that routing changes contribute to end-to-end packet loss significantly. Specifically, we study failover events in which a link failure leads to a routing change and recovery events in which a l...