Measurement of network recovery

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

               Survivability is a measurement of network recovery capability when a failure occurs in the network. It provides connections from source to destination across many network segments, which may have any type of topology or combinations [linear, ring and mesh], besides the network resource discovery and network management connectivity. It defines the principles of the architecture layers for design and represents the components in each layer. It also describes the applications, bringing together the architecture principles such as wire line, cellular and voice over packet networks. In this, network routing is addressed and network elements (NE's) and their functions are described.

               The work of Song Ho-Wu Group [7&8] form the strong foundations for the establishment of technologies and strategies related to survivable fiber optic network architecture. Manonnet Singh Discussed the Survivability parameters in optical networks, using the Hubbing Span Architecture. From the evaluation of Digital Signal Level 3 (DS3) characteristics, Vande.V Discussed the methods of survivability, which can be provided in single fiber network architecture. Starting from the SONET Characteristics, Frank J.Effen Berger , discussed the survivability of SONET in hubbing architecture. Nag T.S Introduced the problem of ad-hoc fair queuing. The focus was how to resolve the conflicts between fairness and maximal throughput. Lisong XU Group depicted the fair channel sharing and limited channel reuse with the generation of Enhanced Maximize -Local-Min Fair Queuing [EMLM-FQ] algorithm. Ju.J Group has carried out the problem with location - dependent contention in ad-hoc network design issues.

               Optical restoration is one of the missing building blocks in today's optical network architecture. In network's survivability, the restoration is not only confined to link but the whole optical segment. To economically utilize the fiber's high capacity, the network architecture often used is organized as a hubbing structure, which is based on a single-homing (SH) concept and/or Dual Homing Concept (DH). Single homing is a centralized demand routing concept that aggregates demands from any office to their destinations through an associated home hub. The fibre-hubbed architecture or single homing architecture is economically attractive, but at the expense of service vulnerability, since a single fiber cut or a hub office failure would isolate a large area served by the Central Office [CO] from communicating with other communities. This can be overcome by Dual Homing [DH] concept which can be evaluated and achieved by Physical Layer and Logical Layer concepts.

               Survivability methodologies are further subdivided in to two types. They are - Traffic Restoration and Facility Restoration. Traffic restoration is applied to switched networks, where as facility restoration is applied to facility transport networks. At present, high-capacity asynchronous fiber facility networks are commonly used is Digital Signal Level (DS3), which carries 45 Mbps of data, rather than Digital Signal Level 0 (DS0), which carries a voice call of 64kbps. Technological advancements play a crucial role in implementing survivable fiber networks.

               Real network contains a mix of protected and unprotected traffic in the network survivability aspect. Some traffic may be protected in other layers. A complete multi-layer and multi-period network optimization has to be considered to yield practical network solutions. A compact cross-connect that interconnects different network architectures in Physical Layer and Logical Layers are discussed.

Description of the research work:

The present work describes the enhancement of node connectivities and their protection mechanisms in terms of TWO LAYERS. They are;

  1. Physical Layer.
  2. Logical Layer.

The Physical Layer is classified into four methodologies. They are;

  1. (Fiber Span Layout and its Demand Distribution, in which the Link Connectivity, Node Connectivity and Digital Cross Connectivity (DCS) factors are measured.
  2. Fiber Network User Service Survivability in which the restoration mechanism is presented, with different types of protection schemes such as 1:1, 1+1, Automatic Protection Switching (APS) and Diverse Protection (DP) are presented.
  3. Optical Network Demand Bundling Using DS3 Forming, in which point-to-point to connections to appropriate DS3 level demands in direct and indirect paths are represented.
  4. SONET (Integrated approach), proposes Multiperiod Synchronous Optical Network Survivability (MSONS) which overcomes the limitations of Hubbing Span Architecture. It assures satisfactory survivability parameter estimations for multi node-to-node connectivity and multiperiod planning of complex network.

               The Physical Layer is enhanced to Logical Layer in which Optical Ad-Hoc network connectivity in terms of Global Fairness model and Maximum Channel Reutilization is presented. It is classified in to three methodologies such as the Two-Tier, Bounded Fair Maximize - Local - Min - Fair Queuing (BFMLM-FQ) and Hybrid Algorithms. The Two-Tier algorithm determines the connectivity for multiperiod fair queuing which achieves the global fairness. It is further extended to BFMLM-FQ algorithm to determine the fairness model by using spatial channel mobility and scalability procedures. Further the Hybrid algorithm describes an integrated model of both Two-Tier and BFMLM-FQ models. These network architectures compared Network Size and Traffic Patterns like Optical Channel Capacity [OCC], Optical Channel Resource Sharing [OCRS], Spatial Locality [SL], Scalability and Node mobility Fairness [SNF & MNF], Multi-hop Optical Network [MOP], and Maximize the Channel Utilization [MCU] in the QoS environment.


               In Physical Layer Fiber Span Layout Demand Distribution, computer results for multilayer survivability parameters are obtained and compared with results available in the literature for Hubbing Span Architecture having 1x5 node connectivity. Numerical results for 1 X 5, 3 X 3, 5 X 5 and 9 X 9 through Point-to- Point Architecture are presented. It is obvious that this method can be easily extended to N X N node connectivity. The DCS Factor by this method for 1 X 5 node connectivity is 80%, where as 20% in the work reported earlier.

               The Fiber Network User Service Survivability (FNUSS) simulates multi restoration parameters and with multiple failures. Results are obtained and compared with those of Hubbing Span Architecture having 1 X 5 node connectivity. The numerical results with single link failure and multiple failures are obtained for 1 X 5 node connectivity in Fiber Network User Service Survivability restoration method. The Survivability Ratio is 20% in Hubbing Span Architecture where as the same is 86% in Point -to -Point Span Architecture. The different types of multi node-to-node connetivities like 3 X 3, 4 X 4, 5 X 5 and 8 X 8 are also simulated which can be extended to N X N connectivity.

               In Optical Network Demand Bundling using Digital Signal 3 Forming, the user connectivity direct path procedure with maximum network traffic propagation connectivity is achieved. The indirect path distributes the demands into different parcel lists through which maximum restoration facility propagation connectivity is obtained.

                In Synchronous Optical Network (SONET), the Multiperiod Service Optical Network Survivability (MSONS) algorithm results in different survivability planning periods. The candidate survivable architecture is achieved for N X N node connectivity and is implemented for different architectures like Fiber Hubbed Span with Diverse Protection (DP) and Point-to-Point Span with DP. Thus maximum and minimum performance of planning period model computation is also achieved.

               In Logical Layer, by using Two-Tier method the global fairness model throughput achieved is 95% for multi-hop network connectivity, as compared to 25% of the previous work. For BFMLM-FQ method the global fairness model throughput achieved is 92% for multihopzconnectivity as compared to 15% of the previous work. Finally, the Hybrid algorithm enhances local and global fairness and also achieves maximized channel utilization throughput of 83.9 and 98.2.


               The entire research work mainly consists of top-down investigations of the benefits of optical layer services and the requirements for the optical layer. A reconfigurable optical network provides faster user connections over the optical network. Automation of such provisioning can be done ultimately where user's equipment and optical network equipment co-operate with each other.

               The planning and designing of survivable optic networks integrates the physical layer model and the logical layer model as computed from various simulation models. Correlation of these results corroborates the validity of the algorithms. The architectures are also presented in the context of incremental changes towards the existing networks in order to improve survivability in optical Networks.

Publications by the Author:

  1. Sairam.K., Dr. T. Janardhana Rao and Dr P.V.D Somasekhar Rao, "Evaluation of Single-Hop Architecture for Metropolitan Area Networks", UBICC Journal 2007, Vol. 2, No 2. P 35-40.
  2. Sairam. K., Dr. T. Janardhana Rao and Dr P.V.D Somasekhar Rao, Optical Communications", UBICC Journal 2007, Vol. 3, No.5, P 26-31.
  3. Sairam. K., Dr. T. Janardhana Rao and Dr P.V.D Somasekhar Rao, "Fiber Optic Survivable Techniques (FOST)", Journal of Institution of Engineers [IE), vol. 87, No.2, July 2006, pp. 25-29.
  4. Sairam. K., Dr. T. Janardhana Rao and Dr P.V.D Somasekhar Rao, "A Packet scheduling Algorithm for Ad-Hoc Optical Networks", IEEE Potentials, Vol.39, No.5, 2006, pp. 30-39.
  5. Sairam. K., "High Performance Optical Networks- An Overview", IJMOT, Vol. 4, No. 1, JANUARY 2009.
  6. Sairam. K., "Efficient Routing Algorithms for Survivability in Optical Networks", UBICC, JANUARY 2009.
  7. Other References:

  8. Song-HO WU. T., David J. Kolar and Richard H. Cardwell, Technologies for Planning a Survivable Fiber Network Architecture. Using Optical Switches", IEEE Journal of Selected Areas in Communications, Vol. 8, No.2, 1990, pp. 152-159.
  9. Song HO-WU. T and Sarry Fouad Habiby, "Survivable Network Architectures for Broadband Fiber Optics", IEEE Transactions on Communications, Vol.40, N0.2, 2003, pp. 125-130.
  10. Manoneet Singh and Subrat Kaur, "Stastical Self-Similarity in Broadband Traffic: Results and Performance Implications", IETE Vol. 17 No.1&2, 2000, pp. 29-36.
  11. Van de .V and Gert Vander Plas," Full Service Optical Access Networks: ATM Transport on Passive Optical Networks", IEEE Communications Magazine, Vol. 41, No.9, 1997, pp. 70-74.
  12. Frank J.Effen Berger, Hhiroshi Ichiubangase and Havro Yamashita, "Performance in Optical Networking Technologies", IEEE Communication Magazine, Vol. 24, No. 5, 2001, pp. 118-132.
  13. Nag T.S., I. Stoica and H. Zhang, "Packet Fair Queuing Algorithms for wireless networks with location-dependent errors," IEEE INFOCOM 98.
  14. Lisong XU, Harry, G. Perros and George Roukas, "Techniques for Optical Packet Switching and Optical Burst Switching", IEEE Communications Magazine", Vol.40 No.5, 2001, pp. 136-142.
  15. Ju. J. and V.O.K. Li, "An Optimal Topology-transparent Scheduling Method in Multihop Packet Radio Networks," IEEE/ACM Transactions on Networking, Vol. 6, No. 3, 1998, pp. 41-49.