This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
But due to the network failure the slow junction of routing protocols has been become serious problem which are growing very fast. Here the term Routing can be defined as a process of finding route or the next hop from source to destination for a packet. Routing is the function of network layer of the OSI model.a device which perform that function is known as routers. Routers use a table known as routing table to find the route of the destination of the packet. This table contains the detail of all the available path of that packet from source to destination; router reads the header of each arriving packet and extracts its detonation address. After extracting the destination address router sends the packet on a best path, which is calculated by considering following points.
Multiple Routing Configuration (MRC) is a procedure which is planned for the recovery in ip network due to the link and node stop working. Multiple Routing configurations are capable to recover link and node failures in IP networks very fastly which is deliberate in milliseconds. MRC used a small set of backup for routing setup. That endorsement helps to desire another path presented in routing table after the failure in the poles apart parts of the system The backup conï¬gurations link weights are set to stay away from routing traffic. Network superintendent monitor that all links which are attached to a node are given adequately high link weights or not. The failure of the node will be able only to affect traffic that is sourced at or destined for the node itself. to exclude a link or a group of links from taking part in the routing, an inï¬nite weight is assigned to it and then That link can be not make the grade without any cost for the traffic.
Multiple Routing Configurations (MRC) is designed into three steps which can be described as
create a set of backup conï¬gurations that can be used by every network component expelled from packet forwarding in one conï¬guration.
for each conï¬guration a average routing algorithm just like Open shortest path first (OSPF) is used, which helps to analyze conï¬guration of shortest paths and for creating forward tables in each router, which is also based on the conï¬gurations. The main advantage of that routing algorithm is that it guarantee loop-free forwarding within one conï¬guration.
Design a forwarding procedure that will make available a advantage of the backup conï¬gurations which will be ready to lend a hand to provide fast recovery from a constituent failure.
Multiple Routing Configurations (MRC) approach, for all links and nodes in the network the backup conï¬gurations is designed, which is configured in the way that link or node is not used to frontward traffic and for every on its own link or node failure, there will be a
offered conï¬guration that will route the traffic to its objective on a path and will also help to avoid the failed element, and for that reason there is designed a valid path with a ï¬nite cost flanked by each node pair.
Multiple Routing Configurations (MRC) allows packet forwarding to keep on over
pre-configured substitute next-hops directly just after the detection of the failure. Multiple Routing Configurations (MRC) provides a fully security of recovery from any single link or node failure, which constitutes a large middle-of-the-road of the failures veteran in a network.
FEATURES OF MULTIPLE ROUTING CONFIGURATIONS (MRC):
â€¢ Multiple Routing Configurations (MRC) gives uninterrupted forwarding of packets in the case of a failure.
â€¢ MRC helps to improve the network availability through containment of the re-convergence process.
â€¢ MRC uses a single method to handle link and node failures. In Multiple Routing Configurations Failures are handled locally by the detecting node, and it always finds a route to the destination if it is equipped.
â€¢ MRC makes no assumptions for the root cause of failure, e.g., whether the packet forwarding is disrupt due to a failed link or a failed router to solve that type of problem MRC guarantees that there exists a valid, preconfigured next-hop to the destination.
â€¢ An performance of MRC can be made without main modifications to existing IGP routing standards.
CONFIGURATION CONSTRAINTS REQUIRED FOR MULTIPLE ROUTING CONFIGURATIONS
Multiple Routing Configurations require a subsequent requirement for the single-failure forbearance and reliable routing, and for the backup configurations
1) A node must not carry any transit traffic in the configuration where it is inaccessible. Still, traffic must be able to depart from and reach an inaccessible node.
2) A link must not carry any traffic in the configuration where it is inaccessible.
3) In each configuration, all node pairs must be connected by a path that path does not pass through an cut off node or an cut off link.
4) Every node and every link must be cut off in at least one backup configuration.
CONï¬GURATIONS STRUCTURE OF MRC ALGORITHM
MRC is connectionless algorithm and its conï¬gurations are deï¬ned by the network topology. All the network topology is same in all conï¬gurations, and the related
Link weights are differ among conï¬gurations. Network topology are represent as a graph G = (N, A), with a set of nodes N and a set of unidirectional links (arcs) A1.
Here the topology graph G must be bi-connected in order to warranty single-fault self-control. A conï¬guration is denied by this topology graph and the connected link weight function. MRC algorithm isolates all nodes in the network, and therefore it requires a bi connected as input. The MRC algorithm can be implemented either in a network management system, or in the routers. And same set of backup conï¬gurations will be calculated As long as all routers have the same vision of the network topology, they will be as follows
Algorithm 1 loops through all nodes in the topology, and tries to cut off them one at a time. A link is also cut off in the same iteration as one of its attached nodes. And when ever either all the nodes and links in the network are cut off in accurately one conï¬guration, or a node that cannot be isolated is encountered, MRC algorithm terminates.
The successful execution of Algorithm 1 results in a complete set of valid backup conï¬gurations.
MRC algorithm executes all the way through all nodes which are trying to make them cut off in one of the backup conï¬gurations .this algorithm can be terminate with or without success. The accomplishment and the failure of the algorithm depend on the following issue.
If the number of conï¬gurations permissible is satisfactorily high then the bi-connected topology will result in a successful termination.
In the case of situation when a node cannot be isolated in any of the conï¬gurations, the algorithm terminates without success.
The term difficulty can be defined as a time or space required to solve a complexity of the MRC algorithm is resolute by the loops and the complexity of the connected method. This method performs a procedure which is similar to determining a node is an expression point in a graph, bound to worst case O (|N|+|A|)
PERFORMANCE EVALUATION of MULIPLE ROUTING CONFIGURATION ALGORITHM
Multiple Routing Configurations (MRC) is a modus operandi which can be defined as a local, proactive recovery scheme designed for the recovery in ip networks due to the link and node failure. MRC algorithm resume packet forwarding without delay after the failure is detected, which results in the fast recovery. In MRC the routers store extra routing configurations and the quantity of utter which is required in the routers, is linked
To the number of backup configurations. In mrc routing in a backup configuration is restricted, but MRC still provides backup paths that are longer than the optimal paths. These Longer backup paths will affect the total network load and also the end-to-end delay. Routing simulator is referred and is also used to evaluate these metrics on a wide range of imitation topologies shortest path routing or "OSPF normal" in the full topology which is chosen as a benchmark for comparison throughout the evaluation. And the purpose of it is to see how close MRC can come within reach of the performance of global OSPF re-convergence.
LOCAL FORWARDING PROCESS OF MRC ALGORITHM
The concept of local forward process in MRC algorithm is based on a one of the standard shortest path algorithm known as open shortest path first algorithm which is used in each configuration. To calculate configuration precise forward tables, here forwarding tables are used to avoid a failed component. As well as a packet arrives at a point of failure, the node which is contiguous to the failure, also known as the detecting node, will be liable for discovery the configuration on which point the failed component is cut off, and to forward the packet according to this configuration. With our proposal, without knowing the basic concept of the failure the detecting node must be able to find the correct configuration. A node must also be able to know in which configuration the downstream node of each of its network interfaces is inaccessible and it must also be known in which configuration it is inaccessible to itself. All these types of information have been provided to the nodes in advance, during the process of configuration generation.
MUTIPLE ROUTING CONFIGURATION EVALUATION Method
1) Routing simulation: to create configurations which are described in the earlier section of the documentation a Java software model has been developed. These are the configurations which are formed for a large range of topologies and these configuration are retrieved from the BRITE topology generation tool (BRITEÂ can be defined as a arguments topology generator that can be used to cram the significance of probable causes for power laws and other metrics which are newly observed in Internet topologies)  using the Waxman  and the Generalized Linear Preference (GLP)  models. The number of nodes is diverse between 16 and 512 to show the scalability. the average node degree is 4 or 6 for Waxman topologies and 3.6 for GLP topologies is assigned and the main purpose of assigning these degree is to explore the effect of network density and For all artificial topologies, the links are specified with a unit weight. In this algorithm for each topology, we measure the minimum number of backup configurations needed to cut off every node and link in the network.
2) Traffic simulation: To test the special effects of this algorithm our scheme has on the weight allocation after a failure, in the implementation of our scheme a discrete-event packet simulator based on the J-Sims framework  has been used. The following figure shows the performance on the European COST239 network  connecting major cities across Europe and All the links have been given a common base weight (dominant), and an individual addition which is based on their propagation delay.
Minimum number of backup configurations:
This figure displays the minimum number of backup configurations in a wide range of synthetic topologies which are required to make all links and nodes inaccessible. In the figure each bar shows 100 different topologies which are provided by the type of generation model used and the links-to-node ratio, and the number of nodes available in the topology.
This table represent the minimum number of mandatory backup Configurations which are needed for five real world topologies. And the result of it displays that the number of backup configurations required is more often than not reserved 3 or 4 because that is normally enough to separate every element in a topology. The number of configurations required will be decreased with the increment of the network connectivity. In him implementation of MRC algorithm user will never required more than six configurations. And this unassuming number of backup configurations represents that this method is implement able without the requirement of a significant amount of state information.
2) Backup path lengths:
This figure displays the path length sharing for all the node failure and the numbers of it are based on 100 different topologies with 32 nodes and 64 links and the Results for link failures and other network properties are displaying the same propensity. Here we have used open shortest path first algorithm in the failure-free case, for distribution of the path length with at least two hops for all paths. For an original path we give permission to each intermediate node fail, and have calculated the resulting backup path lengths using global OSPF rerouting, local rerouting based on the full topology except the failed component ("Optimal local"), as well as MRC with 3 and 7 backup configurations. For this we choose the technique from these samples, and repeat for all paths in the network. and noticed the results that MRC is able to provide the backup path lengths closer to achieve after a full OSPF re-convergence, and to allow the use of more configurations the difference can also be decreased. This means that the affected traffic will not suffer from inappropriately long backup paths in the period when it is forwarded according to an MRC backup configuration.
Link load distribution: this figure explains the link load distribution of traffic results on the basis of MRC algorithm.
This figure displays the link load allocation in the COST239 network (Ultra-high Capacity visual Transmission Networks). This provides the feature of the failure free situation and for OSPF rerouting. The Results are available for OSPF rerouting and MRC using 3 backup configurations. These results are the averages for all 26 achievable link failures. It also provide a mechanism for the link load distribution in the network at the time of when using MRC and after the completion of OSPF re-convergence.
Load on individual links
This figure displays the traffic result on the basis of load on individual links.
This figure displays the weight on each unidirectional link which is available in the network in the failure-free case, and also after a link failure. In the failure-free case these links are indexed from the slightest loaded to the most loaded. These Results are shown for MRC, and will be terminated after the OSPF rerouting procedure. In this concept we have measured the throughput on every link for every possible link failure. Here the Fig. a displays the average for all link failures, while the Fig. b shows the worst case for each creature link and The results displays the average and the worst case both. MRC gives a post-failure load on each link which is comparable to the one achieved after a full OSPF re-convergence. With this case we have also reserved the link weights from the original full topology in the backbone part of the backup topologies.
Total network load:
This figure displays the network load after link failure. Here this metric is linked to the backup path length and also represents the total traffic loads in the network after a failure. The subordinate-most favorable backup paths given by MRC should be able to result in an amplified load in the network. Here the Fig. 7 displays the cumulative throughput of all the links in the COST239 network after a link failure (Ultra-high Capacity visual Transmission Networks). Here the x-axis displays which of the 26 bidirectional links has failed. And the y-axis displays the relatively increased load compared to the failure-free case and with it, it also displays that the load in the network has been increased about 5% on average after a failure at the time of using MRC with 3 backup configurations, in the comparison to a 2% increase with OSPF rerouting. Here All traffic is recovered in this picture, which results to the increased network load which is exclusively caused by the longer paths experienced by the rerouted traffic.
GENERATING BACKUP CONFIGURATIONS
For generate backup configuration from the link and node failure here first of all we have gathered the requirements that must be put on the backup configurations used in MRC. Then, we suggest an algorithm that can be easily used to mechanically create such configurations. MRC algorithm will naturally be run once at the preliminary startup of the network, and after that it will be run each time when a node or link is enduringly added or removed in a network. In the following section we have described the technique which will provide a help in a generating backup configuration:-
Here we can add 4 points which I have send u in the attachment dec21
In this section the first condition decides what amount of weights must be put on the classified links which are attached to an unreachable node. To make available an assurance hat no path will go through an inaccessible node, it suffices that the restricted links have a weight W which will not be a less than the sum of all link weights w which are defined in the innovative configuration:
This will also provide the assurance of that several other path which are available between two nodes in the network will be chosen by a shortest path algorithm (open shortest path algorithm) before one passing through the inaccessible node. And In this section the second condition will implies that the weight of an inaccessible link must be set in the way so that traffic will never be routed above it. These links are provided an inestimable weight. The purpose of Giving these boundaries on the link weights are that we will now move on to show that how we can create backup configurations that remain to the last two requirements which are acknowledged above.