A Method with Node Autonomy to Preserve QoS in Networks with Per-Route Admission

 

A. Mateos-Papis

 

División de Ciencias de la Comunicación y Diseño. Universidad Autónoma Metropolitana. Unidad Cuajimalpa. México, DF., México.

 

ABSTRACT

This paper presents a method called the Short Term Protection (STP) which is applied to fixed networks where flows¹ are constrained to follow specific routes, where the admission-process is distributed, per-route, and where the traffic is sensitive to delay (like MPEG4 traffic). The share of bandwidth of any route is protected, in the short-term, against the traffic-increment of its intersecting-routes. In the long term, in every congested link, the share of bandwidth between the routes that intersect at that link is proportional to the average relative measured demands of those routes. The nodes act autonomously, without central administration. This method is expected to: 1) help network administrators to have confidence, within a time-frame , about the amount of bandwidth they count on for every route; 2) allow a simple bandwidth management in the network, with prospect to be scalable to at least tens of nodes. This paper presents the general and detailed operation of the method, the evaluation of the method, by simulations, including a comparison with other method, a discussion of a possible scenario of application, conclusions and possible paths for further research.

Keywords: quality of service (QoS); bandwidth sharing; distributed traffic control.

 

RESUMEN

Este artículo presenta un método llamado short term protection -STP- (Protección a Corto Plazo) que se aplica a redes fijas donde los flujos¹ están restringidos a seguir rutas específicas, donde el proceso de admisión es distribuido, por ruta, y donde el tráfico es sensible a retrasos (como el tráfico MPEG4). La porción de ancho de banda de cada ruta está protegida, en el corto plazo, contra incremento de tráfico en rutas de intersección. En el largo plazo, en cada enlace congestionado, la proporción de ancho de banda entre las rutas que se interceptan en ese enlace es proporcional a las demandas relativas promedio medidas de esas rutas. Los nodos actúan de manera autónoma, sin administración central. Se espera que este método: 1) Ayude a los administradores de la red a tener confianza, por un periodo de tiempo, acerca de la cantidad de ancho de banda con la que cuentan para cada ruta; 2) Permita una administración simple en la red, con prospecto de ser escalable hacia al menos decenas de nodos. Este artículo presenta la operación general y detallada del método, la evaluación del método, por simulaciones, incluyendo una comparación con otro método, un diálogo sobre un posible escenario de aplicación, conclusiones y posibles rutas para investigación futura.

 

DESCARGAR ARTÍCULO EN FORMATO PDF

 

References

[1] V. Kumar et al, "Router Architectures for the Differentiated Services of Tomorrows Internet", IEEE Communications Magazine, Vol. 36, No. 5, pp. 152-164, 1998.         [ Links ]

[2] S. Blake et al, "An Architecture for Differentiated Services", RFC 2475, IETF, 1998.         [ Links ]

[3] K. Nichols et al, "Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers", RFC 2474, IETF, 1998.         [ Links ]

[4] C. Cetinkaya and E. Knightly, "Egress Admission Control", in: 19th IEEE International Conference on Computer Communications, INFOCOM 2000, The Conference on Computer Communications, Tel Aviv, Israel, 2000, pp. 1471-1480.         [ Links ]

[5] C. Cetinkaya et al. "Scalable Services via Egress Admission Control", IEEE Transactions on Multimedia: Special Issue on Multimedia over IP, Vol. 3, No. 1, pp. 69-81, 2001.         [ Links ]

[6] T. Hui and C. Tham, "Adaptive Provisioning of Differentiated Services Networks Based on Reinforcement Learning", IEEE Transactions on Systems, Man, and Cybernetics, Vol. 33, No. 4, pp. 492-501, 2003.         [ Links ]

[7] C. Li and E. Knightly, "Schedulability Criterion and Performance Analysis of Coordinated Schedulers", IEEE/ACM Transactions on Networking, Vol. 13, No. 2, pp. 276-287, 2005.         [ Links ]

[8] M. F. Horng et al, "An Adaptive Approach to Weighted Fair Queue with QoS Enhanced on IP Network", in: IEEE Region 10 International Conference on Electrical and Electronic Technology, (TENCON 2001), Singapore, Singapore, 2001, pp. 181-186.         [ Links ]

[9] H. Wang et al, "Adaptive-Weighted Packet Scheduling for Premium Service", in: IEEE International Conference on Communications, ICC 2001, Helsinki, Finland, 2001, pp. 1846-1850.         [ Links ]

[10] D. Hang et al, "TD2FQ: An Integrated Traffic Scheduling and Shaping Scheme for DiffServ Networks", in: IEEE Workshop on High Performance Switching and Routing", Dallas, Texas, USA, 2001, pp. 78-82.         [ Links ]

[11] C. C. Li et al, "Proportional Delay Differentiation Service Based on Weighted Fair Queuing", in: IEEE Ninth International Conference on Computer Communications and Networks, ICCCN 2000, Las Vegas, Nevada, USA, 2000, pp. 418-423.         [ Links ]

[12] S. Floyd and V. Jacobson V., "Random Early Detection Gateways for Congestion Avoidance", IEEE/ACM Transactions on Networking, Vol. 1, No. 4, pp. 397-413, 1993.         [ Links ]

[13] A. Parekh and R. Gallager, "A Generalized Processor Sharing Approach to Flow Control in Integrated Services Networks: The Single Node Case", IEEE/ACM Transactions on Networking, Vol. 1, No. 3, pp. 346-357, 1993.         [ Links ]

[14] ns-2 Network Simulator, Information Science Institute, 2011, Marina del Rey, California, USA, Available from: http://www.isi.edu/nsnam/ns/        [ Links ]

[15] E. Altman and T. Jimenez, "NS Simulator for Beginners, Lecture Notes, 2003-2004", Univ. de Los Andes, Merida, Venezuela and ESSI, Sophia-Antipolis, France. Available from: http://www-sop.inria.fr/maestro/personnel/Eitan.Altman/COURSNS/n3.pdf.         [ Links ]

[16] P. Pieda et al, "A Network Simulator Differentiated Services Implementation", Open IP, Nortel Networks. 2000. Available from: http://www-sop.inria.fr/members/Eitan.Altman/COURSNS/DOC/DSnortel.pdf.         [ Links ]

[17] A. Mrkaic, "Porting a WFQ Scheduler into ns-2's Diffserv Environment", Student Thesis, Computer Engineering and Networks Laboratory, Swiss Federal Institute of Technology, 2001.         [ Links ]

[18] J. L. Devore, "Probability and Statistics for Engineering and Sciences". The 6th Edition. Brooks Cole, ISBN 0534-39933-9. ISBN 13: 978-0534-39933-7. 2003.         [ Links ]

[19] A. Matrawy et al, "MPEG4 Traffic Modeling Using the Transform Expand Sample Methodology", in: IEEE 4th International Workshop on Networked Appliances, Gaithersburg, Maryland, USA, 2002, pp. 249-256. Also available, 2012, from: "ns-2 contributed-code", http://nsnam.isi.edu/nsnam/index.php/Contributed_Code #Topology_and_Traffic_Generation        [ Links ]

[20] G. Salmon et al, "NetFPGA-Based Precise Traffic Generation", in: NetFPGA Developers Workshop'09, Standford University, California, USA, 2009.         [ Links ]

[21] CIDR REPORT, 2012. Available from: http://www.cidr-report.org/as2.0/        [ Links ]

[22] E. Nygren et al, "The Akamai Network: A Platform for High-Performance Internet Applications", ACM SIGOPS Operating Systems Review, Vol. 44(3), pp. 2-19, 2010.         [ Links ]

[23] Analyzing UDP usage in Internet traffic. The Cooperative Association for Internet Data Analysis. 2012. Available from: http://www.caida.org/research/trafficanalysis/tcpudpratio/        [ Links ]

[24] C. Lee et al, "Unmasking the growing UDP traffic in a Campus Network", in: Passive and Active Measurement (PAM) Conference, Vienna, Austria, 2012.         [ Links ]

[25] E. Kohler and S. Floyd, "Datagram Congestion Control Protocol (DCCP) Overview", ICSI Center for Internet Research, Berkeley, CA, USA, 2003.         [ Links ]

[26] J. Huang et al, "An Effective Approximation Scheme for Multiconstrained Qualiy-of-Service Routing", in: IEEE Global Telecommunications Conference (GLOBECOM 2010), pp. 1-6, 6-10, 2010.         [ Links ]

[27] G. Xue et al, "Finding a Path Subject to Many Additive QoS Constraints", IEEE/ACM Transactions on Networking, Vol. 15, No. 1, pp. 201-211, 2007.         [ Links ]

[28] Y. Xiao et al. "Computing a Most Probable Delay Constrained Path: NP-Hardness and Approximation Schemes", IEEE Transactions on Computers, Vol. 61, No. 5, pp. 738-744, 2012.         [ Links ]

 

Nota

¹ A flow is a sequence of related packets that enter a network through the same source-node and leave the network through the same egress-node.