Network sensitivity to hot-potato disruptions

Teixeira, Renata; Shaikh, Aman; Griffin, Tim; Voelker, Geoffrey M.

doi:10.1145/1030194.1015493

Cited by 24 publications

(26 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hot-potato shifts can also be caused by link failures [23,14]. A link failure can only increase the cost of paths between two points that used that link, since IS-IS or OSPF will now have to pick a more expensive path around it.…”

Section: Impact Of More Frequent Changesmentioning

confidence: 99%

See 1 more Smart Citation

Measuring the Shared Fate of IGP Engineering and Interdomain Traffic

Agarwal

Nucci²,

Bhattacharyya

13TH IEEE International Conference on Network Protocols (ICNP'05)

View full text Add to dashboard Cite

Typically, each Autonomous System (AS) tunes its local IS-IS or OSPF metrics without any coordination with otherASes. Such local optimizations can lead to sub-optimal end-to-end network performance, as suggested by the performance enhancements achieved by some overlay routing projects. We study the interaction of local IGP engineering in an ISP network with interdomain routing policies. Specifically, (a) how does hot-potato routing (the BGP policy of choosing the closest egress) influence the selection of IGP link metrics? and (b) how does traffic to neighboring ASes shift due to changes in the local AS's IGP link metrics?In our measurement study, we find that the hot-potato routing policy interacts significantly with IGP engineeringignoring this interaction resulted in metrics sub-optimal by as much as 20% of link utilization. Further, the impact on neighboring ASes depends on peering locations and policies, and as much as 25% of traffic to a neighboring AS can shift the exit point. Such interdomain shifts can be detrimental to the performance of neighboring ASes. We rely on the actual measured network topology, IGP metrics, traffic matrix and delay bounds. Even though our results are specific to a single ISP, they show significant interaction between local IGP engineering and interdomain routing policies, and thus motivate further work on global network optimization and coordination among ISPs.

show abstract

Section: Impact Of More Frequent Changesmentioning

confidence: 99%

“…Hot-potato variations can occur during typical network operation [1,24]. Teixeira et al [23] have examined how these variations occur when the IGP topology changes. While closely related, their focus is on network failure episodes, such as link failures, router failures and forwarding loops.…”

Section: Related Workmentioning

confidence: 99%

Measuring the Shared Fate of IGP Engineering and Interdomain Traffic

Agarwal

Nucci²,

Bhattacharyya

13TH IEEE International Conference on Network Protocols (ICNP'05)

View full text Add to dashboard Cite

show abstract

“…The selection of next hop for a prefix follows a specified procedure where the BGP-specific attributes are examined followed by a comparison of distance to next-hop according to the intradomain routing protocol. In many cases, the intradomain routing decides exit point for the prefix, as the operator typically tries to select the closest exit point according to the intradomain routing cost (a policy often referred to as hot potato routing [31]). A more detailed description of BGP4 can be found in [17].…”

Section: Routing and Load Balancing In The Internetmentioning

confidence: 99%

“…Many large traffic shifts that occur in a network are due to changes in the interdomain routing or interactions between inter and intradomain routing [31][32][33]. On the one hand, changes in interdomain routing parameters influence the selection of next hop by BGP, as exaplained in Sect.…”

Section: Sources Of Traffic Variationsmentioning

confidence: 99%

Robust load balancing under traffic uncertainty—tractable models and efficient algorithms

Gunnar

Johansson

2010

Telecommun Syst

View full text Add to dashboard Cite

Routing configurations that have been optimized for a nominal traffic scenario often display significant performance degradation when they are subjected to real network traffic. These degradations are due to the inherent sensitivity of classical optimization techniques to changes in model parameters combined with the significant traffic variations caused by demand fluctuations, component failures and network reconfigurations. In this paper, we review important sources for traffic variations in data networks and describe tractable models for capturing the associated traffic uncertainty. We demonstrate how robust routing settings with guaranteed performance for all foreseen traffic variations can be effectively computed via memory efficient iterative techniques and polynomial-time algorithms. The techniques are illustrated on real data from operational IP networks.

show abstract

“…The work in [18] proposed metrics of network sensitivity to measure the hot-potato disruptions. TIE mechanism is proposed to tune route egress selection, which allows an ingress router to have a different ranking for different destination prefixes.…”

Section: Related Workmentioning

confidence: 99%

AS Egress Selection Optimization Based On Link States

Liu

Zhang

Zhao

et al. 2006

2006 14th IEEE International Conference on Networks

View full text Add to dashboard Cite

This paper presents an optimization mechanism to facilitate efficient selection of BGP egress for an AS, when IGP link state changes. Most mechanisms for BGP egress selection have primarily focused on achieving an optimized goal by considering it as a static problem, which leads to inflexibility to variations in the topology of an AS, especially for link failures which often occur in an AS network. Our mechanism provides a flexible means for an AS to optimize routing according to their multiple goals. Based on precomputed control rules and current link state, every BGP router can select appropriate egress points by itself. The mechanism is extensible, flexible, and robust. A critical algorithm based on link states is applied to illustrate the main idea of the mechanism. Simulation results demonstrate that our solution is feasible and expressive for the network administrators.

show abstract

Network sensitivity to hot-potato disruptions

Cited by 24 publications

References 17 publications

Measuring the Shared Fate of IGP Engineering and Interdomain Traffic

Measuring the Shared Fate of IGP Engineering and Interdomain Traffic

Robust load balancing under traffic uncertainty—tractable models and efficient algorithms

AS Egress Selection Optimization Based On Link States

Contact Info

Product

Resources

About