100GET-E3 R3G

funded this research project. Part of the CELTIC initiative.

Project Team

Heads	Prof. Dr.-Ing. Phuoc Tran-Gia, Dr. Michael Menth, Dr. Rastin Pries
Researchers	Michael Duelli
Students	Julian Ott, Vlad Singeorzan, Thilo Müller
Former Students	Christopher Pluntke, Eduard Weber, Xiao-Hua Qin, Anke Endler
Funded period	October 2007 - September 2010, October 2010 - December 2010

Summary

Data communication demands for more and more capacity. DSL access of 16 Mbit/s is common in Germany, 100 Mbit/s is already offered by some providers. Fiber access is state of the art in many countries of South-East Asia and tremendous traffic volumes are exchanged over fiber lines. This trend will continue over the next decade. The market is competitive such that Internet service providers (ISPs) need to provide high-quality services (high throughput, little delay, little jitter, high reliability) at low cost.

The CELTIC initiative brings together university researchers and business partners to jointly face this challenge. ADVA, Alcatel-Lucent, Ericsson, and Nokia-Siemens Networks are the main drivers each of them heading a group of university and small or medium enterprise partners. The Chair of Distributed Systems of the University of Wuerzburg is part of the subproject "100GET-E3: End-to-End Carrier-Grade Ethernet" led by Nokia-Siemens Networks and receives funding under the name "R3G: Routing, Grooming, Resilience, and Resource Management for Multilayer Networks".

The objective of the project 100GET is to provide a low-cost high-capacity solution for the network of the future. Two main objectives are addressed: the transmission of 100 Gbit/s and the provision of high-quality services at low cost. The project R3G focuses on the second aspect.

One of the main cost factors for ISPs are network components and among them IP routers are most expensive. Therefore, ISPs look for alternatives that are cheaper with respect to investment and operation cost. Enhanced Ethernet technology - the so-called Carrier Ethernet - seems to be a candidate to replace many IP routers within networks of ISPs. Therefore, international standardization bodies like IEEE or IETF push forward T-MPLS, VLAN XC, and PBB-TE/PBT.

The research of the University of Wuerzburg targets at the following issues. Cost savings can be obtained by keeping traffic in the optical domain as long as possible, using packet-switching capabilities of Carrier Ethernet nodes inside the network only when needed, and banning IP routers from the core. This is achieved by intelligent traffic grooming and routing that should make best use of the network resources. In addition, the reliability of the network is improved by cost-efficient protection switching and restoration mechanisms. To that end, optimization methods for the design and configuration of a future communication architecture are developed.

Project Partners

Cooperating business partners:	Cooperating university partners:

On the Web

EUREKA SUCCESS STORY: CELTIC 100GET (2011-06-10)
Collaborative research projects win Gold Excellence Award at Celtic-Plus Event 2011 (2011-03-30)
Weltrekord in serieller Datenübertragung via Glasfaser – 10,2 Tbit/s oder 240 DVDs in einer Sekunde (2011-03-09)
100GET and all included sub-projects receive the Celtic Gold Award! (2011-03-02)
Verizon links France and Germany with 100G Ethernet - Paris to Frankfurt backbone upgraded (2010-11-12)
100G-Technologie im Deutschen Forschungsnetz (DFN) (2010-09-14)
Cisco schließt Übernahme von CoreOptics ab (2010-07-15)
Standard für 100-GBit/s-Ethernet kommt gerade recht (2010-06-24)
Glasfaserbetrieb mit 100 GBit/s auf einer Wellenlänge (2010-06-17)
Optische 100-GBit/s-Übertragungstechnik in Silizium (2010-03-12)
Datenturbo aus dem Labor (2010-03-09)
Halbzeit für das 100GET-Projekt (2009-07-20)
Stauwarnung für den Transatlantik-Datenverkehr (2009-06-24)
Cisco: Verfünffachung des Internet-Verkehrs bis 2013 (2009-06-16)
Juniper bringt erste 100-GE-Schnittstelle für Enterprise-Router (2009-06-09)
100-Gigabit-Ethernet zeigt sich tauglich für die Praxis (2008-07-07)
100GET-E3 Project Leaflet (PDF) (2008-06-20)
Spiegel.de: Article on the use of holograms for future, economic communication (2008-01-21)
Celtic Press Release: More bandwidth and security for the Internet, New Celtic project 100GET develops 100 Gigabit Ethernet (2008-01-11)
BMBF Pressemitteilung: "Europäische Innovationsallianz für das Internet der Zukunft gestartet" (2007-12-11)
Spiegel.de: Article on the Possible Collapse of the Internet in 2010 (2007-11-21)

Publications

Greedy Design of Resilient Multi-Layer Networks Duelli, Michael (2010, June 4).
- [ BibTeX ]
- [ EndNote ]
@presentation{info3-talk-2010-375, address = {6th Conference on Next Generation Internet Networks (NGI 2010), Paris, France}, author = {Duelli, Michael}, keywords = {100get}, month = 6, title = {Greedy Design of Resilient Multi-Layer Networks}, year = 2010 }

%0 Generic %1 info3-talk-2010-375 %A Duelli, Michael %C 6th Conference on Next Generation Internet Networks (NGI 2010), Paris, France %D 2010 %T Greedy Design of Resilient Multi-Layer Networks
Greedy Design of Resilient Multi-Layer Networks Duelli, Michael; Qin, Xiaohua; Menth, Michael in 6th Conference on Next Generation Internet Networks (NGI 2010) (2010).
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
In this paper, we propose a deterministic greedy heuristic providing a construction layout for a cost-efficient multi-layer network that is able to carry a given set of traffic demands with and without protection on different layers. We apply the heuristic to different reference network topologies and protection requirements. Evaluations are conducted regarding equipment cost on different layers, blocking probability, path lengths, and number of demands affected by specific failures.

@inproceedings{info3-inproceedings-2010-382, abstract = {In this paper, we propose a deterministic greedy heuristic providing a construction layout for a cost-efficient multi-layer network that is able to carry a given set of traffic demands with and without protection on different layers. We apply the heuristic to different reference network topologies and protection requirements. Evaluations are conducted regarding equipment cost on different layers, blocking probability, path lengths, and number of demands affected by specific failures.}, address = {Paris, France}, author = {Duelli, Michael and Qin, Xiaohua and Menth, Michael}, booktitle = {6th Conference on Next Generation Internet Networks (NGI 2010)}, keywords = {100get}, month = 6, title = {Greedy Design of Resilient Multi-Layer Networks}, year = 2010 }

%0 Conference Paper %1 info3-inproceedings-2010-382 %A Duelli, Michael %A Qin, Xiaohua %A Menth, Michael %B 6th Conference on Next Generation Internet Networks (NGI 2010) %C Paris, France %D 2010 %T Greedy Design of Resilient Multi-Layer Networks %X In this paper, we propose a deterministic greedy heuristic providing a construction layout for a cost-efficient multi-layer network that is able to carry a given set of traffic demands with and without protection on different layers. We apply the heuristic to different reference network topologies and protection requirements. Evaluations are conducted regarding equipment cost on different layers, blocking probability, path lengths, and number of demands affected by specific failures.
Deferring CAPEX Investments in Multi-Layer Networks through IP Traffic-Dependent Expansion Stages Duelli, Michael; Endler, Anke; Menth, Michael in 11. ITG-Fachtagung Photonische Netze (2010). 133–138.
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
Network design is performed for a trafﬁc load in the future. This trafﬁc load is far greater than trafﬁc loads observed at intermediate stages. In this paper, we propose to plan a multi-layer network for such a future trafﬁc load, but instead of provisioning the full network equipment at once, we suggest to defer the deployment of expensive IP interfaces and other enabling equipment only to stages when they are really needed to carry trafﬁc. We develop an algorithm that concentrates IP trafﬁc at an intermediate stage on a subset of those IP interfaces that are required to carry the full load in the future. This reduces the set of used IP interfaces which need to be installed at intermediate stages, and CAPEX investments for the other equipment can be deferred to the future.

@inproceedings{info3-inproceedings-2010-374, abstract = {Network design is performed for a trafﬁc load in the future. This trafﬁc load is far greater than trafﬁc loads observed at intermediate stages. In this paper, we propose to plan a multi-layer network for such a future trafﬁc load, but instead of provisioning the full network equipment at once, we suggest to defer the deployment of expensive IP interfaces and other enabling equipment only to stages when they are really needed to carry trafﬁc. We develop an algorithm that concentrates IP trafﬁc at an intermediate stage on a subset of those IP interfaces that are required to carry the full load in the future. This reduces the set of used IP interfaces which need to be installed at intermediate stages, and CAPEX investments for the other equipment can be deferred to the future.}, address = {Leipzig, Germany}, author = {Duelli, Michael and Endler, Anke and Menth, Michael}, booktitle = {11. ITG-Fachtagung Photonische Netze}, keywords = {100get}, month = 5, pages = {133–138}, title = {Deferring CAPEX Investments in Multi-Layer Networks through IP Traffic-Dependent Expansion Stages}, year = 2010 }

%0 Conference Paper %1 info3-inproceedings-2010-374 %A Duelli, Michael %A Endler, Anke %A Menth, Michael %B 11. ITG-Fachtagung Photonische Netze %C Leipzig, Germany %D 2010 %P 133--138 %T Deferring CAPEX Investments in Multi-Layer Networks through IP Traffic-Dependent Expansion Stages %X Network design is performed for a trafﬁc load in the future. This trafﬁc load is far greater than trafﬁc loads observed at intermediate stages. In this paper, we propose to plan a multi-layer network for such a future trafﬁc load, but instead of provisioning the full network equipment at once, we suggest to defer the deployment of expensive IP interfaces and other enabling equipment only to stages when they are really needed to carry trafﬁc. We develop an algorithm that concentrates IP trafﬁc at an intermediate stage on a subset of those IP interfaces that are required to carry the full load in the future. This reduces the set of used IP interfaces which need to be installed at intermediate stages, and CAPEX investments for the other equipment can be deferred to the future.
A Generic Algorithm for CAPEX-Aware Multi-Layer Network Design Duelli, Michael (2009, May 5).
- [ BibTeX ]
- [ EndNote ]
@presentation{info3-talk-2009-336, address = {10. ITG-Fachtagung Photonische Netze, Leipzig, Germany}, author = {Duelli, Michael}, keywords = {100get}, month = 5, title = {A Generic Algorithm for CAPEX-Aware Multi-Layer Network Design}, year = 2009 }

%0 Generic %1 info3-talk-2009-336 %A Duelli, Michael %C 10. ITG-Fachtagung Photonische Netze, Leipzig, Germany %D 2009 %T A Generic Algorithm for CAPEX-Aware Multi-Layer Network Design
Network Planning, Control and Management Perspectives on Dynamic Networking Michaelis, Thomas; Duelli, Michael; Chamania, Mohit; Lichtinger, Bernhard; Rambach, Franz; Türk, Stefan in European Conference and Exhibition on Optical Communication (ECOC) (2009).
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
Besides the frequently claimed cost savings due to automation, one major advantage of dynamic networking is that this networking paradigm matches the mode of operation which is observed in the Internet and is essential for making transport networks a native part of it. This talk highlights three essential ingredients of dynamic networking: Firstly, network planning creates the required degrees of freedom in the data plane for a dynamic control plane. Secondly, service provisioning and restoration provide the required agility for dynamic network operation, no matter if implemented on the management or on a control plane. Eventually, service management ensures adequate service performance across all involved network domains. We conclude that all of these aspects need to be addressed adequately in order to realize dynamic networking.

@inproceedings{info3-inproceedings-2009-356, abstract = {Besides the frequently claimed cost savings due to automation, one major advantage of dynamic networking is that this networking paradigm matches the mode of operation which is observed in the Internet and is essential for making transport networks a native part of it. This talk highlights three essential ingredients of dynamic networking: Firstly, network planning creates the required degrees of freedom in the data plane for a dynamic control plane. Secondly, service provisioning and restoration provide the required agility for dynamic network operation, no matter if implemented on the management or on a control plane. Eventually, service management ensures adequate service performance across all involved network domains. We conclude that all of these aspects need to be addressed adequately in order to realize dynamic networking.}, address = {Vienna, Austria}, author = {Michaelis, Thomas and Duelli, Michael and Chamania, Mohit and Lichtinger, Bernhard and Rambach, Franz and Türk, Stefan}, booktitle = {European Conference and Exhibition on Optical Communication (ECOC)}, keywords = {100get}, month = 9, title = {Network Planning, Control and Management Perspectives on Dynamic Networking}, year = 2009 }

%0 Conference Paper %1 info3-inproceedings-2009-356 %A Michaelis, Thomas %A Duelli, Michael %A Chamania, Mohit %A Lichtinger, Bernhard %A Rambach, Franz %A Türk, Stefan %B European Conference and Exhibition on Optical Communication (ECOC) %C Vienna, Austria %D 2009 %T Network Planning, Control and Management Perspectives on Dynamic Networking %X Besides the frequently claimed cost savings due to automation, one major advantage of dynamic networking is that this networking paradigm matches the mode of operation which is observed in the Internet and is essential for making transport networks a native part of it. This talk highlights three essential ingredients of dynamic networking: Firstly, network planning creates the required degrees of freedom in the data plane for a dynamic control plane. Secondly, service provisioning and restoration provide the required agility for dynamic network operation, no matter if implemented on the management or on a control plane. Eventually, service management ensures adequate service performance across all involved network domains. We conclude that all of these aspects need to be addressed adequately in order to realize dynamic networking.
A Generic Algorithm for CAPEX-Aware Multi-Layer Network Design Duelli, Michael; Weber, Eduard; Menth, Michael in 10. ITG-Fachtagung Photonische Netze (2009). 167–174.
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
Capital expenditure (CAPEX), i.e. equipment cost, is a decisive criterion for the design of a new network infrastructure. The design of efficient transport networks requires the interconnection of multiple technologies which form separate layers with their own networking view and routing. This leads to the notion of multi-layer networks. CAPEX-aware design of such networks requires algorithms which propose multi-layer hardware configurations that are able to carry a given set of traffic demands at minimal CAPEX. We present a generic algorithm for CAPEX-aware multi-layer network design as well as a computationally viable implementation and perform evaluations on realistic network topologies. The underlying CAPEX and multi-layer technology models are explained in detail.

@inproceedings{info3-inproceedings-2009-343, abstract = {Capital expenditure (CAPEX), i.e. equipment cost, is a decisive criterion for the design of a new network infrastructure. The design of efficient transport networks requires the interconnection of multiple technologies which form separate layers with their own networking view and routing. This leads to the notion of multi-layer networks. CAPEX-aware design of such networks requires algorithms which propose multi-layer hardware configurations that are able to carry a given set of traffic demands at minimal CAPEX. We present a generic algorithm for CAPEX-aware multi-layer network design as well as a computationally viable implementation and perform evaluations on realistic network topologies. The underlying CAPEX and multi-layer technology models are explained in detail.}, address = {Leipzig, Germany}, author = {Duelli, Michael and Weber, Eduard and Menth, Michael}, booktitle = {10. ITG-Fachtagung Photonische Netze}, keywords = {100get}, month = 5, pages = {167–174}, title = {A Generic Algorithm for CAPEX-Aware Multi-Layer Network Design}, year = 2009 }

%0 Conference Paper %1 info3-inproceedings-2009-343 %A Duelli, Michael %A Weber, Eduard %A Menth, Michael %B 10. ITG-Fachtagung Photonische Netze %C Leipzig, Germany %D 2009 %P 167--174 %T A Generic Algorithm for CAPEX-Aware Multi-Layer Network Design %X Capital expenditure (CAPEX), i.e. equipment cost, is a decisive criterion for the design of a new network infrastructure. The design of efficient transport networks requires the interconnection of multiple technologies which form separate layers with their own networking view and routing. This leads to the notion of multi-layer networks. CAPEX-aware design of such networks requires algorithms which propose multi-layer hardware configurations that are able to carry a given set of traffic demands at minimal CAPEX. We present a generic algorithm for CAPEX-aware multi-layer network design as well as a computationally viable implementation and perform evaluations on realistic network topologies. The underlying CAPEX and multi-layer technology models are explained in detail.
CAPEX-Aware Design of Survivable DWDM Mesh Networks Pluntke, Christopher; Menth, Michael; Duelli, Michael in IEEE International Conference on Communications (ICC) (2009).
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
This paper reviews the basic architecture and component costs of opaque, transparent, and semi-transparent DWDM networks and looks at the network design problem from a capital expenditure (CAPEX) point of view. Given are a ﬁber topology and a demand matrix with different bit rates. Required is the least-cost optical equipment for that topology together with the routing and potential muxponder-based aggregation of all demands such that they can be supported by the newly designed network. We look at the problem for networks without resilience requirements and for survivable networks using 1+1 protection against single ﬁber cuts. We model this problem for the three types of optical networks by integer linear programs (ILPs) in a canonical way.

@inproceedings{info3-inproceedings-2009-342, abstract = {This paper reviews the basic architecture and component costs of opaque, transparent, and semi-transparent DWDM networks and looks at the network design problem from a capital expenditure (CAPEX) point of view. Given are a ﬁber topology and a demand matrix with different bit rates. Required is the least-cost optical equipment for that topology together with the routing and potential muxponder-based aggregation of all demands such that they can be supported by the newly designed network. We look at the problem for networks without resilience requirements and for survivable networks using 1+1 protection against single ﬁber cuts. We model this problem for the three types of optical networks by integer linear programs (ILPs) in a canonical way.}, address = {Dresden, Germany}, author = {Pluntke, Christopher and Menth, Michael and Duelli, Michael}, booktitle = {IEEE International Conference on Communications (ICC)}, keywords = {100get}, month = 6, title = {CAPEX-Aware Design of Survivable DWDM Mesh Networks}, year = 2009 }

%0 Conference Paper %1 info3-inproceedings-2009-342 %A Pluntke, Christopher %A Menth, Michael %A Duelli, Michael %B IEEE International Conference on Communications (ICC) %C Dresden, Germany %D 2009 %T CAPEX-Aware Design of Survivable DWDM Mesh Networks %X This paper reviews the basic architecture and component costs of opaque, transparent, and semi-transparent DWDM networks and looks at the network design problem from a capital expenditure (CAPEX) point of view. Given are a ﬁber topology and a demand matrix with different bit rates. Required is the least-cost optical equipment for that topology together with the routing and potential muxponder-based aggregation of all demands such that they can be supported by the newly designed network. We look at the problem for networks without resilience requirements and for survivable networks using 1+1 protection against single ﬁber cuts. We model this problem for the three types of optical networks by integer linear programs (ILPs) in a canonical way.
Performance Evaluation of IP over Cost-Optimized Optical Multilayer Networks with SRLGs Duelli, Michael (2008, April 28).
- [ BibTeX ]
- [ EndNote ]
@presentation{info3-talk-2008-271, address = {9. ITG-Fachtagung Photonische Netze, Leipzig, Germany}, author = {Duelli, Michael}, keywords = {100get}, month = 4, title = {Performance Evaluation of IP over Cost-Optimized Optical Multilayer Networks with SRLGs}, year = 2008 }

%0 Generic %1 info3-talk-2008-271 %A Duelli, Michael %C 9. ITG-Fachtagung Photonische Netze, Leipzig, Germany %D 2008 %T Performance Evaluation of IP over Cost-Optimized Optical Multilayer Networks with SRLGs
Minimizing Installation Costs of Survivable DWDM-Mesh Networks: A Heuristic Approach Duelli, Michael; Pluntke, Christopher; Menth, Michael in 4th Euro-NGI Conference on Next Generation Internet Networks (NGI 2008) (2008). 15–22.
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
The cost function for the capacity of optical links follows a step function. That means, the support of one more lightpath might require a costly upgrade of an optical cross connect (OXC), but then additional lightpaths can be supported at almost no further cost. This should be considered when lightpaths are routed through an optical network. In this paper we optimize the routing of the lightpaths to minimize the costs for the required optical equipment. We consider this problem for the failure-free case only and for survivable networks using dedicated path protection. We formulate the problems by integer linear programs (ILPs). In addition, we propose heuristics to solve the problem since solving ILPs is computationally expensive and not feasible for large problem instances. We show that our heuristics lead to good results within a fraction of time compared to ILP solvers.

@inproceedings{info3-inproceedings-2008-311, abstract = {The cost function for the capacity of optical links follows a step function. That means, the support of one more lightpath might require a costly upgrade of an optical cross connect (OXC), but then additional lightpaths can be supported at almost no further cost. This should be considered when lightpaths are routed through an optical network. In this paper we optimize the routing of the lightpaths to minimize the costs for the required optical equipment. We consider this problem for the failure-free case only and for survivable networks using dedicated path protection. We formulate the problems by integer linear programs (ILPs). In addition, we propose heuristics to solve the problem since solving ILPs is computationally expensive and not feasible for large problem instances. We show that our heuristics lead to good results within a fraction of time compared to ILP solvers.}, address = {Krakow, Poland}, author = {Duelli, Michael and Pluntke, Christopher and Menth, Michael}, booktitle = {4th Euro-NGI Conference on Next Generation Internet Networks (NGI 2008)}, keywords = {100get}, month = 4, pages = {15–22}, title = {Minimizing Installation Costs of Survivable DWDM-Mesh Networks: A Heuristic Approach}, year = 2008 }

%0 Conference Paper %1 info3-inproceedings-2008-311 %A Duelli, Michael %A Pluntke, Christopher %A Menth, Michael %B 4th Euro-NGI Conference on Next Generation Internet Networks (NGI 2008) %C Krakow, Poland %D 2008 %P 15--22 %T Minimizing Installation Costs of Survivable DWDM-Mesh Networks: A Heuristic Approach %X The cost function for the capacity of optical links follows a step function. That means, the support of one more lightpath might require a costly upgrade of an optical cross connect (OXC), but then additional lightpaths can be supported at almost no further cost. This should be considered when lightpaths are routed through an optical network. In this paper we optimize the routing of the lightpaths to minimize the costs for the required optical equipment. We consider this problem for the failure-free case only and for survivable networks using dedicated path protection. We formulate the problems by integer linear programs (ILPs). In addition, we propose heuristics to solve the problem since solving ILPs is computationally expensive and not feasible for large problem instances. We show that our heuristics lead to good results within a fraction of time compared to ILP solvers.
Performance Evaluation of IP over Cost-Optimized Optical Multilayer Networks with SRLGs Duelli, Michael; Hartmann, Matthias; Menth, Michael; Hülsermann, Ralf; Düser, Michael in 9. ITG-Fachtagung Photonische Netze (2008). 21–29.
- [ Abstract ]
- [ BibTeX ]
- [ EndNote ]
- [ Download ]
The design of optical transport and IP networks is investigated with respect to the effect on overall network perfor mance when the two aspects are combined in the same network architecture. The focus is on the identiﬁcation of advantages in the joint design of IP and transport networks, as well as on the shortcomings and network bottlenecks which may result from sub-optimal design. The impact of different IP routing strategies (shortest path, equal-cost multipath) on the network utilization is quantiﬁed, and potential sources for overload in the IP domain are determined. Simulations were carried out to systematically assess technical metrics like link utilization and economic parameters such as cost for a 50-node transport network with line rates up to 40 Gbit/s. The impact of ﬁber cuts was taken into account, leading to multiple logical IP links to fail simultaneously, i.e. shared risk link groups (SRLGs), and triggering massive IP rerouting. Applying the results allows to identify bottlenecks in the design, and to devise mechanisms which allow cost-optimal network design of future IP and optical transport networks.

@inproceedings{info3-inproceedings-2008-312, abstract = {The design of optical transport and IP networks is investigated with respect to the effect on overall network perfor mance when the two aspects are combined in the same network architecture. The focus is on the identiﬁcation of advantages in the joint design of IP and transport networks, as well as on the shortcomings and network bottlenecks which may result from sub-optimal design. The impact of different IP routing strategies (shortest path, equal-cost multipath) on the network utilization is quantiﬁed, and potential sources for overload in the IP domain are determined. Simulations were carried out to systematically assess technical metrics like link utilization and economic parameters such as cost for a 50-node transport network with line rates up to 40 Gbit/s. The impact of ﬁber cuts was taken into account, leading to multiple logical IP links to fail simultaneously, i.e. shared risk link groups (SRLGs), and triggering massive IP rerouting. Applying the results allows to identify bottlenecks in the design, and to devise mechanisms which allow cost-optimal network design of future IP and optical transport networks.}, address = {Leipzig, Germany}, author = {Duelli, Michael and Hartmann, Matthias and Menth, Michael and Hülsermann, Ralf and Düser, Michael}, booktitle = {9. ITG-Fachtagung Photonische Netze}, keywords = {100get}, month = 4, pages = {21–29}, title = {Performance Evaluation of IP over Cost-Optimized Optical Multilayer Networks with SRLGs}, year = 2008 }

%0 Conference Paper %1 info3-inproceedings-2008-312 %A Duelli, Michael %A Hartmann, Matthias %A Menth, Michael %A Hülsermann, Ralf %A Düser, Michael %B 9. ITG-Fachtagung Photonische Netze %C Leipzig, Germany %D 2008 %P 21--29 %T Performance Evaluation of IP over Cost-Optimized Optical Multilayer Networks with SRLGs %X The design of optical transport and IP networks is investigated with respect to the effect on overall network perfor mance when the two aspects are combined in the same network architecture. The focus is on the identiﬁcation of advantages in the joint design of IP and transport networks, as well as on the shortcomings and network bottlenecks which may result from sub-optimal design. The impact of different IP routing strategies (shortest path, equal-cost multipath) on the network utilization is quantiﬁed, and potential sources for overload in the IP domain are determined. Simulations were carried out to systematically assess technical metrics like link utilization and economic parameters such as cost for a 50-node transport network with line rates up to 40 Gbit/s. The impact of ﬁber cuts was taken into account, leading to multiple logical IP links to fail simultaneously, i.e. shared risk link groups (SRLGs), and triggering massive IP rerouting. Applying the results allows to identify bottlenecks in the design, and to devise mechanisms which allow cost-optimal network design of future IP and optical transport networks.

back to NGN group

100GET-E3 R3G

100GET-E3 R3G

Project Team

Summary

Project Partners

On the Web

Publications

Data privacy protection

Data privacy protection

Picture credits