Experience in measuring internet backbone traffic variability: Models metrics, measurements and meaning

Roughan, Matthew; Greenberg, Albert; Kalmanek, Charles; Rumsewicz, M.; Yates, Jennifer; Yin, Zhang

doi:10.1016/s1388-3437(03)80183-x

Cited by 99 publications

(129 citation statements)

References 7 publications

Supporting

Mentioning

126

Contrasting

Order By: Relevance

“…The two largest annotated networks in this database are beyond the computational capabilities of our mixed-integer optimization formulation. To generate a synthetic traffic matrix from the graphs provided by Rocketfuel, we use a gravity model [22] in a manner similar to Applegate and Cohen [13] where the inferred link weights are used to calculate approximate bandwidths of each link at a node. These bandwidths are used in the gravity model to derive the proportion of traffic for which each node (PoP) is responsible and therefore how much traffic it sends to all the other nodes.…”

Section: Test Network and Traffic Matricesmentioning

confidence: 99%

Power Awareness in Network Design and Routing

Chabarek

Sommers

Barford

et al. 2008

2008 Proceedings IEEE INFOCOM - The 27th Conference on Computer Communications

166

251

View full text Add to dashboard Cite

Abstract-Exponential bandwidth scaling has been a fundamental driver of the growth and popularity of the Internet. However, increases in bandwidth have been accompanied by increases in power consumption, and despite sustained system design efforts to address power demand, significant technological challenges remain that threaten to slow future bandwidth growth. In this paper we describe the power and associated heat management challenges in today's routers. We advocate a broad approach to addressing this problem that includes making powerawareness a primary objective in the design and configuration of networks, and in the design and implementation of network protocols. We support our arguments by providing a case study of power demands of two standard router platforms that enables us to create a generic model for router power consumption. We apply this model in a set of target network configurations and use mixed integer optimization techniques to investigate power consumption, performance and robustness in static network design and in dynamic routing. Our results indicate the potential for significant power savings in operational networks by including power-awareness.

show abstract

Section: Test Network and Traffic Matricesmentioning

confidence: 99%

Power Awareness in Network Design and Routing

Chabarek

Sommers

Barford

et al. 2008

2008 Proceedings IEEE INFOCOM - The 27th Conference on Computer Communications

166

251

View full text Add to dashboard Cite

show abstract

“…For GÉANT, its traffic matrices were collected on May 5th, 2005 for every 15 minutes; we obtained both the topology and traffic matrices from the authors of [11]. For Sprint and AT&T, we randomly generate a traffic matrix using the gravity model [12], and scaled the traffic to obtain 40 different traffic matrices. Simulation runs were carried out on a Linux PC with 3.07GHz CPU and 8GB RAM.…”

Section: A Experiments Setupmentioning

confidence: 99%

Power-aware routing in networks with delay and link utilization constraints

Lin

Soh

Lazarescu

et al. 2012

37th Annual IEEE Conference on Local Computer Networks

View full text Add to dashboard Cite

Current network infrastructures are over-provisioned and thus exhibit poor power efficiency at low traffic load. In this paper, we consider networks comprising of bundled links, whereby each link has one or more physical cables that can be switched off independently. The problem at hand is then to switch off redundant cables during off peak periods, while retaining the QoS provided to existing traffic demands. Unfortunately, the problem to maximally shutdown redundant cables is an NP-complete problem. Henceforth, we design a fast heuristic, called Multiple Paths by Shortest Path First (MSPF), that aims to maximize the number of switched-off cables subject to satisfying maximum link utilization (MLU) and end-to-end delay requirements. We have extensively evaluated the performance of MSPF on both real and synthetic topologies and traffic demands. Further, we have compared its performance against two state-of-the-art techniques: GreenTE usable only when each link has one cable, and FGH that supports bundled links but usable only for networks without MLU and delay constraints. MSPF improves the energy saving on average by 5% as compared to GreenTE incurring only 1% the CPU time. While yielding equivalent energy savings, MSPF requires only 0.35% of the running time of FGH. Finally, for MLU at most 50% and end-to-end delay no longer than the network diameter, MSPF reduces the power usage of the GÉANT topology up to 91% and bundled links consisting of ten cables.

show abstract

“…For Abilene, we select a subclass of the online available matrices [16], which are measured on March 1st, 2004. The traffic matrix for Cernet2 is generated by the gravity model [15] based on aggregated link load collected from January 10th to 16th, 2010. We generate a sparse traffic matrix for Random through gravity model to achieve a light traffic load scheme.…”

Section: Methodsmentioning

confidence: 99%

Routing On Demand: Toward the Energy-Aware Traffic Engineering with OSPF

Shen

Liu

et al. 2012

Networking 2012

View full text Add to dashboard Cite

Abstract.Energy consumption has already become a major challenge to the current Internet. Most researches aim at lowering energy consumption under certain fixed performance constraints. Since trade-offs exist between network performance and energy saving, Internet Service Providers (ISPs) may desire to achieve different Traffic Engineering (TE) goals corresponding to changeable requirements. The major contributions of this paper are twofold: 1) we present an OSPF-based routing mechanism, Routing On Demand (ROD), that considers both performance and energy saving, and 2) we theoretically prove that a set of link weights always exists for each trade-off variant of the TE objective, under which solutions (i.e., routes) derived from ROD can be converted into shortest paths and realized through OSPF. Extensive evaluation results show that ROD can achieve various trade-offs between energy saving and performance in terms of Maximum Link Utilization, while maintaining better packet delay than that of the energy-agnostic TE.

show abstract

Experience in measuring internet backbone traffic variability: Models metrics, measurements and meaning

Cited by 99 publications

References 7 publications

Power Awareness in Network Design and Routing

Power Awareness in Network Design and Routing

Power-aware routing in networks with delay and link utilization constraints

Routing On Demand: Toward the Energy-Aware Traffic Engineering with OSPF

Contact Info

Product

Resources

About