Multiple routing configurations for Fast IP Network Recovery have the slow convergence of routing protocol. It presented the new scheme called Multiple Routing Configurations (MRC). MRC is connectionless and it assumes the destination based as hop-by-hop. Network of IP are intrinsically robust and IGP protocols are like OSPF that to design and update the changed topology information after the failure. MRC is proactive and allows the recovery in range of milliseconds with protection mechanism. MRC has the main idea to use the network graph and links that associate the weights to produce network configuration.
IP network recovery:
According to Amund Kvalbein, Audun Fosselie Hansen, Tarik Cicic, the IP networks are intrinsically robust. The IGP protocols like OSPF are design to update the forward information based on the topology which is changed after the failure, when the new information state is distributed to each router to calculate new routing tables. The IP re-convergence is the time -consuming process and a link is difficult to find the period of routing. During this packets can be dropped due to invalid routes. Both IGP and BGP context are studied and has the adverse effect on real-time application. Many efforts are optimize in different steps to convergence of IP routing as detection, dissemination of information and to calculate the shortest part.
2.1 IP Fast network failure recovery:
According to Mingui Zhang, Bin Liu and Beichuan Zhang, in the IP level the routers of the network still should able to be deliver the packets in a long alternative path that exist. The perfect routing protocols are Open Shortest Path First (OSPF) and Intermediate System to Intermediate System (IS-IS). These routers are informed on network topology to change the update messages and re-calculate the routing path. The different approaches to handle the physical failures are Proactive Failure Recovery (PFR) were the routers compute and store the backup paths for potential failures and once the local link failure is detected. PFR has the shortest failure recovery time and can reduce both the update propagation and path re-calculation. The large portion of failures in IP network is short, transient failure and PFR should able to improve the quality of deliver packets. The goals for minimizing the failures of application performances use the post-failure load balance which is important. Weight-setting method is used for load balance in PFR. It gives the demand of the projected from previous measurements to find the weight set to avoid congestions. The links weights are re-assigned according to the load that carriers the large weights for heave loaded links and small weights for light loaded links and re-calculate the routing path based on the new link weights. PFR can has two design challenges of post failure load balancing that is first is how to select the multiple loop-free backup path with small and second is how to decide the amount of affected traffic that is to shed on each backup path in balanced manner. By proposing the unique solutions can follow the four contributions as 1) explore and identify other two additional types of loop-free alternative paths that is beside the Equal-Cost-Multiple-path that gives more path diversity to post-failure delivery,2) decides the traffic distribution over multiple paths that formulate it as LP problem that minimize the sum of links which are utilize. To solve the problem, assign the penalty factor to heavy loaded link to achieve the goal to maximize and minimize the utilized links in the network. 3) Introduce the base factor into LP iteration to damp the oscillation. 4) If the failure occurs, the LP problem can solve incrementally where the traffic affected by the failure is taken in consideration by objective function of LP. The result shows the simulation that the scheme has effectively balance to load the multiple paths which has small computation and fast converges.
2.2 Post-failure Load Balancing:
It describes how to divert the traffic to backup paths so that causing of risk congestions can be minimized. In this there are different ways for post failure load balancing as ODL matrix, formulation of linear programming, LP Iteration and incremental IP Iterations (Mingui Zhang, Bin Liu and Beichuan Zhang).
2.2.1 ODL Matrix:
The matrix of IP network can represent by Origination-Destination matrix (OD matrix).Suppose OD (j, k) were (j, k= 1, 2… N; j not equal to k) denotes the traffic volume from origin node j to destination node k. N is the total number of nodes in the network. The load of each link can be calculated from the data structure ODL through the manipulation of special matrix elements.
2.2.2 Formulation of liner programming:
In this the weight -setting approach will formulate the traffic allocation problem as ILP problem. It is formulated in different LP problem which has more tractable. It establish the multiple alternative paths and has the details that encoded by the data structure of ODL. The weight-setting method has to restore to solve the ILP. Here IP is simple than the IPL. If LP is finished the backup paths get proportion of traffic according to their capability.
2.2.3 LP Iteration:
The objective function of LP which treats the contribution of each link which is utilized to overall cost equally and objective function of optimization is not true. The equation of the objective function is used arbitrarily without any analysis and is still complex to be used in n-objective functions were it is piece wise linear functions rather than the linear functions to establish the contribution link.
2.2.4 Incremental LP Iteration:
The working paths of the traffic effects by the failure can be shed the alternative path locally. If there is no alternative path for failure recovery, the upstream router can send the Internet Control Message Protocol (ICMP) to redirect the message to the source. The shifting of the diverted traffic can be done incrementally with the help of ODL matrix that has the details easily gotten by the affected traffic. Re-allocate the affected based traffic with the current traffic by LP, when LP is completed the failure is recovered and effected traffic is shifted to backup path. Incremental IP iteration has the best to avoid the congestions that are not to disturb the unaffected traffic begins carried by the network. The speed of the work is compared with weight-setting method as replace ILP with light-weighted LP, the damps oscillations of the iteration converges are quickly and balance the diverted load without resorting the time consuming to re-calculate the routing path in the network (Mingui Zhang, Bin Liu and Beichuan Zhang).
2.3 Securing IP network storage:
According to Charles Baumert, storage area network are developed by Fiber Channel technologies (FC). The network FC can provide the high performance of connection to storage the dedicated links. Many different technologies are provides the ability to interconnect the SANs using Conventional transmission control protocol/Internet Protocol (TCP/IP) and Ethernet network, these include fiber channel internet protocol (FCIP), internet fiber channel protocol (iFCP) and internet small computer system interface (iSCSI). The advantage of using this IP is that the new technologies can establish the security methods as IP Security (IPSec) as, encryption, authentication and data integrity. Encryption keeps the important information confidential private. Authentication can ensure the identities of both the sender and receiver to communicate and to authentic before the information is exchanged and data integrity is used to integrate the information during transmission (Charles Baumert).
2.4 Technologies for securing IP network:
The IP storage network requires the security of local network segment. Additional security will be provided by segregating the storage resources, restricting the access on the base of operating system and shared storage access. The IP network can do with confidence to establish the IPSec protocol. In tunneling mode the IP traffic is issued from the individual host that is encrypted and then wrapped additional IP address to the remote security gateway. By the end of the IPSec tunnel the traffic is decrypted and the original header is used to route the traffic in final destination. The capabilities of VPN security can add the other network equipment as routers. The high performance and security are been provided with encryption device that can separately managed by the routers (Charles Baumert).
2.5 Load balancing:
It ensures the system availability in the event with the usage of the system. In load-balance configuration multiple NetHawks are been deployed parallel. The router with load balancing has capability to divide the amount of processing the work between VPN gateway and the router so that the processing is optimize by all the users that are served quickly. The several types of load balancing can include stationary, round-robin and load measure to ensure the scalability and flexibility (Charles Baumert).
3. Multiple Routing Configurations (MRC):
To assure fast recovery from link and node failures in IP networks, we present a new recovery scheme called Multiple Routing Configurations (MRC). The proposed scheme guarantees recovery in all single failure scenarios, using a single mechanism to handle both link and node failures, and without knowing the root cause of the failure.MRC is strictly connectionless, and assumes only destination based hop-by-hop forwarding.
The main idea of MRC is to use the network graph and the associated link weights to produce a small set of backup network configurations. The link weights in these backup Configurations are manipulated so that for each link and node failure, and regardless of whether it is a link or node failure, the node that detects the failure can safely forward the incoming packets towards the destination on an alternate link. MRC assumes that the network uses shortest path routing and destination based hop-by-hop forwarding.
According to Amund Kvalbein, Multiple Routing Configurations (MRC) is a proactive and local protection mechanism that allows recovery in the range of milliseconds. MRC is based on building a small set of backup routing configurations that are used to route traffic on alternate paths after a failure. The observation is that if all links attached to a particular node are given sufficiently high link weights, traffic will never be routed through that node. The failure of that node will then affect traffic that acts as a source node at or destined for the node itself.
According to T.Cicic, the Internet takes an increasingly central role in our communications infrastructure; the slow convergence of routing protocols after a network failure becomes a growing problem. To assure fast recovery from link and node failures in IP networks, we present a new recovery scheme called Multiple Routing Configurations (MRC). MRC is based on keeping additional routing information in the routers, and allows packet forwarding to continue on an alternative output link immediately after the detection of a failure. It can be implemented with only minor changes to existing solution. In present MRC, and analyze its performance with respect to scalability, backup path lengths, and load distribution after a failure. We also show how an estimate of the traffic demands in the network can be used to improve the distribution of the recovered traffic, and thus reduce the chances of congestion when MRC is used.
3.1 Using multiple routing protocols:
According to Tom Lancaster, multiple routing protocols can implement in different ways on the network. Protocols not even aware to each other and also the information can not exchange. It uses the Interior Gateway Protocol (IGP) and Exterior Gateway Protocol (EGP). It clearly understands how to accomplish the same sets of two protocols that are used in two different ways. It plans to allow the packets to pass the network by source and destination. The network administrator often that, whether it is easier to configure the redistribution between IGP and EGP and controls the network with access-lists. Mutual redistribution is a good idea in enterprise network. The network statements commands are static and simple to understand the result in deterministic behavior and there is no need to guess what network is going to do. Multiple paths are useful to exit the network and have the good balance between dynamic and controlled. It remembers that two protocols are used by giving the flexibility of intelligent decision-making in EGP without sending routes into IGP and by default filters are much useful.
3.2 Static route analysis based multiple route configurations:
According to Hartinder Singh Johal, Dynamic routing protocol had dominated the network thorough out the world. The protocols like RIP, OSPF, IS-IS that dynamically adapt the network topology to change with low management in vast array of network based applications. In the same way static routing faces the task to maintain its credibility to dynamic routing. The less static routing has inherent the advantages as simple configuration and easy predication with robustness of network problems like routing convergence failure and external route spoofing. Static routing is not really a protocol and they do not maintain the distributed algorithm for route calculation in case of dynamic routing protocol. The application demands to retain the benefits of static routing and asks for some degree of flexibility and fault tolerance by dynamic routing. The intend to explore the probability that integrate the degree of fault tolerance to static routing by multiple alternative routes that has similar concepts of backup and floating static routes that sorts on basic optimality principle using sink tree.
3.3 Routing Supports:
TCP/IP supports static and dynamic routing, if we use any routing the mechanism is same. TCP/IP routes packet are searched by its routing table. It supports multiple equal-cost routes to maintain the equal cost paths that allow the size of TCP/IP virtual machine. Static routing uses the gateway statements in profile TCPIP to define the static routers. The routers do not change the response automatically. If the destination is down through static route, the static route remains in routing table and traffic continues to send towards the destination without any success. The dynamic routing in the table are, entities are dynamically managed and can automatically changes the response to network topology. These routers are managed by routing daemon, MPRoute. MPRoute is recommended routing daemon for TCP/IP. It is an IP routing daemon to support RIP version 1, and 2. It makes the use of its own configuration files and do not use the gateway configuration. Configure the subnet mask, destination address and other parameters with OSPF_INTERFACE, RIP_INTERFACE statements in MPRoute configuration file and also we can configure MTU value for an interface by using OSPF_INTERFACE, RIP_INTERFACE by specifying the MTU values on the MTU option to link the statement in profile TCPIP.
3.4 IP forwarding and routing:
IP provides the delivery service as end-to-end for IP datagram's between the hosts. The services are realized with help of IP routers. The IP routers forward the datagram in hop-by-hop fashion from source to destination.
3.4.1 End -to-end delivery of IP datagram:
It composed the collection of networks that are in LAN, a switched network or single point to point link. The internet protocol provides to connect the multiple networks to form single internetwork, which each pair of host can be connected to internetwork that can exchange the IP datagram's. The systems that are connected with the network called IP routers and the router has two or more network interface that are connected in different networks and also forwards the IP datagram between the connected networks. The data link layer of the network is exactly corresponds to one IP network. End-to-end delivery of IP datagram has the success with the relation between the data link networks and the networks of IP must be satisfied with the set of conditions. The delivery of IP datagram, routers can involves in two distinct processes that is it process refers the reception of datagram to lookup the IP destination address in the routing table and transmission the datagram to the destination. The other process is to calculate the paths which determine the contents of routing table. This routing can be static or dynamic. In static routing, the entities of the routing tables are manually configured but in dynamic routing the calculation of route can performed by routing protocol. Routing protocol implements the distributed algorithm where to exchange the routers information to calculate the best paths between the hosts and also uses the shortest path algorithm.
3.5 Debugging static routing configurations:
In this protocol analyzer as ethereal, the commands ping, trace route are the main tools for debugging the static routing setup. The ping command does not provide the information where the failure occurred. The trace route tool can use to determine the part of the path which is correctly configured. The routing table of the last router displays the trace route which uses the incorrect routing table. By capturing ICMP traffic with ethereal, it provide the useful hints on configuration error of static routing. The entry of the routing table is missed in destination, the ICMP network of unreachable message is send to source.
Methodologies:
A great deal of work has recently been done to get better robustness against constituent failures in IP networks Internet is also becoming increasingly central role in communication. The routings of the multiple protocols becomes slow due to network failures. It will be growing to a big problem for all internet users. In network there may be number of nodes and links. Data can be transfer through the nodes from source to origin. Recovering the network failures can give the efficient transformation on IP networks.
A lot of effort has been committed to optimizing the dissimilar steps of the union of IP routing that is discovery, distribution of information and shortest path computation, but the union time is still very huge for applications.
Gantt chart:
