www.ijecs.

in
International Journal Of Engineering And Computer Science ISSN:2319-7242
Volume2 Issue 8 August, 2013 Page No. 2628-2634

Multiple Routing Configurations For Fast Ip Network

Recovery
Sk.Meeravali1 B.Bobby Arun Kumar2 N.Naveen Kumar3 Nakka Thirupathi Rao4 Kuda
Nageswara Rao5
Department of IT, Sir C.R.R.E.C, Eluru, A.P, India
Department of CSE, Scient Institute of Technology, Andhra Pradesh
Department of CSE, Scient Institute of Technology, Andhra Pradesh
Department of CS&SE, A.U. College of Engineering (A) , AU, Visakhapatnam
Department of CS&SE , A.U. College of Engineering (A), AU, Visakhapatnam

Abstract:
As 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). Our 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. 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
solutions. In this

paper we 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.

Key words: TCP/IP, IGP, OSPF Routing protocols, communication services. The demands on Internet
MRC, MRC System.
reliability and availability have increased
1. INTRODUCTION accordingly. A disruption of a link in central parts
In recent years the Internet has been
of a network has the potential to affect hundreds
transformed from a special purpose network to a
ubiquitous platform for a wide range of everyday
Sk.Meeravali1 IJECS Volume 2 Issue 8 August, 2013 Page No.2628-2634 Page 2628
of thousands of phone conversations or TCP normal IP convergence process can be put on hold.
connections, with obvious adverse effects. This process is then initiated only as a consequence
The ability to recover from failures has of non-transient failures. Since no global re-routing
always been a central design goal in the Internet. IP is performed, fast failure detection mechanisms like
networks are intrinsically robust, since IGP routing fast hellos or hardware alerts can be used to trigger
protocols like OSPF are designed to update the MRC without compromising network stability.
forwarding information based on the changed MRC guarantees recovery from any single link or
topology after a failure. This re-convergence node failure, which constitutes a large majority of
assumes full distribution of the new link state to all the failures experienced in a network. MRC makes
routers in the network domain. When the new state no assumptions with respect to the root cause of
information is distributed, each router individually failure, e.g., whether the packet forwarding is
calculates new valid routing tables. disrupted due to a failed link or a failed router.
This network-wide IP re-convergence is a 1.2 MODULES
time consuming process, and a link or node failure 1.2.1 TOPOLOGY CONSTRUCTION:
is typically followed by a period of routing In this module, we construct a topology
instability. During this period, packets may be structure. Here we use mesh topology because of its
dropped due to invalid routes. This phenomenon has unstructured nature. Topology is constructed by
been studied in both IGP and BGP context, and has getting the names of the nodes and the connections
an adverse effect on real-time applications. Events among the nodes as input from the user. While
leading to a re-convergence have been shown to getting each of the nodes, their associated port and
occur frequently. ip address is also obtained. For successive nodes,
The IGP convergence process is slow the node to which it should be connected is also
because it is reactive and global. It reacts to a accepted from the user. While adding nodes,
failure after it has happened, and it involves all the comparison will be done so that there would be no
routers in the domain. In this paper we present a node duplication. Then we identify the source and
new scheme for handling link and node failures in the destinations.
IP networks. Multiple outing Configurations (MRC) 1.2.2 MESSAGE TRANSMISSON:
is a proactive and local protection mechanism that In this module we transmit the message
allows recovery in the range of milliseconds. MRC from source to destination. Here we choose a
allows packet forwarding to continue over destination and select a shortest path for that
preconfigured alternative next-hops immediately destination. Shortest path is calculated by Dijkstra
after the detection of the failure. Using MRC as a Algorithm. it will take minimum node cost an
first line of defense against network failures, the account to find the path between a source and

Sk.Meeravali1 IJECS Volume 2 Issue 8 August, 2013 Page No.2628-2634 Page 2629
destination. The shortest path is updated in the We distinguish between the normal
routing table. The source obtains the shortest path configuration and the backup configurations, Ci, i >
from the routing table itself. After receiving a 0. In the normal configuration, all links have
message the destination will send an ―normal‖ weights W0(a) Є {1…Wmax}. We
acknowledgement to the corresponding source. assume C0 that is given with finite integer weights.
1.2.3 PREVENTING LINK FAILURE USING MRC is agnostic to the setting of these weights. In
MRC: the backup configurations, selected links and nodes
Our MRC approach is threefold. First, we must not carry any transit traffic. Still, traffic must
create a set of backup configurations, so that every be able to depart from and reach all operative nodes.
network component is excluded from packet Isolated links do not carry any traffic.
forwarding in one configuration. Restricted links are used to isolate nodes from
Second, for each configuration, a standard traffic forwarding. The restricted link weight must
routing algorithm like OSPF is used to calculate be set to a sufficiently high, finite value to achieve
configuration specific shortest paths and create that. Nodes are isolated by assigning at least the
forwarding tables in each router, based on the restricted link weight to all their attached links.
configurations. 1.2.4 LOAD DISTRIBUTION:
The use of a standard routing algorithm The shifting of traffic to links bypassing the
guarantees loop-free forwarding within one failure can lead to congestion and packet loss in
configuration. Finally, we design a forwarding parts of the network. This limits the time that the
process that takes advantage of the backup proactive recovery scheme can be used to forward
configurations to provide fast recovery from a traffic before the global routing protocol is
component failure. informed about the failure, and hence reduces the
In our approach, we construct the backup chance that a transient failure can be handled
configurations so that for all links and nodes in the without a full global routing re-convergence.
network, there is a configuration where that link or Ideally, a proactive recovery scheme should
node is not used to forward traffic. not only guarantee connectivity after a failure, but
Thus, for any single link or node failure, also do so in a manner that does not cause an
there will exist a configuration that will route the unacceptable load distribution.
traffic to its destination on a path that avoids the With MRC, the link weights are set
failed element. Also, the backup configurations individually in each backup configuration. This
must be constructed so that all nodes are reachable gives great flexibility with respect to how the
in all configurations, i.e., there is a valid path with a recovered traffic is routed. The backup
finite cost between each node pair. configuration used after a failure is selected based

Sk.Meeravali1 IJECS Volume 2 Issue 8 August, 2013 Page No.2628-2634 Page 2630
on the failure instance, and thus we can choose link the network uses shortest path routing and
weights in the backup configurations that are well destination based hop-by-hop forwarding.
suited for only a subset of failure instances. The shifting of traffic to links bypassing the
2.1 EXISTING SYSTEM: failure can lead to congestion and packet loss in
IP networks are intrinsically robust, since parts of the network. This limits the time that the
IGP routing protocols like OSPF are designed to proactive recovery scheme can be used to forward
update the forwarding information based on the traffic before the global routing protocol is
changed topology after a failure. Much effort has informed about the failure, and hence reduces the
been devoted to optimizing the different steps of chance that a transient failure can be handled
the convergence of IP routing, i.e., detection, without a full global routing re-convergence.
dissemination of information and shortest path Ideally, a proactive recovery scheme should not
calculation, but the convergence time is still too only guarantee connectivity after a failure, but also
large for applications with real time demands. do so in a manner that does not cause an
Disadvantages: unacceptable load distribution. This requirement
This network-wide IP re-convergence is a has been noted as being one of the principal
time consuming process and a link or node failure challenges for recalculated IP recovery schemes.
is typically followed by a period of routing With MRC, the link weights are set individually in
instability. During this period, packets may be each backup configuration. This gives great
dropped due to invalid routes. flexibility with respect to how the recovered traffic
The IGP convergence process is slow is routed. The backup configuration used after a
because it is reactive and global. It reacts to a failure is selected based on the failure instance, and
failure after it has happened. For the existing thus we can choose link weights in the backup
system global routing information is needed. configurations that are well suited for only a subset
2.2 PROPOSED SYSTEM: of failure instances.
The main idea of MRC is to use the network Advantages:
graph and the associated link weights to produce a Multiple Routing Configurations (MRC) is a
small set of backup network configurations. The proactive and local protection mechanism that
link weights in these backup configurations are allows recovery in the range of milliseconds.
manipulated so that for each link and node failure, MRC allows packet forwarding to continue
and regardless of whether it is a link or node over preconfigured alternative next-hops
failure, the node that detects the failure can safely immediately after the detection of the failure.
forward the incoming packets towards the
destination on an alternate link. MRC assumes that

Sk.Meeravali1 IJECS Volume 2 Issue 8 August, 2013 Page No.2628-2634 Page 2631
 Using MRC as a first line of defense against  User clicks on Send button to initiate the
network failures, the normal IP convergence Message transmission in MRC System.
process can be put on hold.  MRC System then checks the Shortest path
3 Module Implementation from the Paths Table
3.1 Topology Construction: This flow diagram  Then, the MRC System checks whether the
provides the flow for Topology Creation in MRC selected shortest path really exists or not?
System. The sequence of steps are provided below  If the shortest path exists, Message is
 A Node is entered by the User using the transmitted on that path
Java Swing UI Front end  Otherwise, an alternative shortest path is
 Upon entering the node information, the calculated and message is transmitted along
system checks whether the node is present in that path.
the NodeInfo table or not? 3.4 Load Distribution: This flow diagram provides
 If the node is already present on NodeInfo, the flow for Load Distribution in MRC System. The
do nothing. Otherwise, sequence of steps are provided below
 Add the node to NodeInfo table.  User provides a node to be logged in.
3.2 Message Transmission: This flow diagram  Then the system will check the
provides the flow for Node Login in MRC corresponding links to that particular node
System. The sequence of steps are provided from Links Table
below  If the node is isolated, load to that node will
 User enters a Node to be logged in as. This be blocked.
will be the source node  Otherwise, load to that node will be allowed.
 Then, the user selects the destination node to Thus, load is balanced in MRC System.
where the message needs to be transferred 4. SIMULATION ANALYSIS

 With the Source Node and Destination

Node, the MRC System computes the
shortest path. This will make use of Paths
Table
 Then, the message is transferred along the
shortest path from Source to Destination.
3.3 Preventing Failure Using MRC: This flow Fig 4.1 SITE-TO-SITE VPN
diagram provides the flow for Preventing Failure
using MRC System. The sequence of steps are
provided below
Sk.Meeravali1 IJECS Volume 2 Issue 8 August, 2013 Page No.2628-2634 Page 2632
 6. CONCLUSIONS
In this paper, Multiple Routing
Configurations as an approach to achieve fast
recovery in IP networks is proposed. MRC is based
on providing the routers with additional routing
configurations, allowing them to forward packets
along routes that avoid a failed component. MRC
guarantees recovery from any single node or link
failure in an arbitrary bi-connected network. By
calculating backup configurations in advance, and
operating based on locally available information
only, MRC can act promptly after failure discovery.
MRC operates without knowing the root
cause of failure, i.e., whether the forwarding
disruption is caused by a node or link failure. This
is achieved by using careful link weight assignment
according to the rules we have described. The link
weight assignment rules also provide basis for the

Sk.Meeravali1 IJECS Volume 2 Issue 8 August, 2013 Page No.2628-2634 Page 2633
specification of a forwarding procedure that [4] C. Boutremans, G. Iannaccone, and C. Diot,
successfully solves the last hop problem. The ―Impact of link failures on VoIP performance,‖ in

performance of the algorithm and the forwarding Proc. Int. Workshop on Network and Operating
System Support for Digital Audio and Video, 2002,
mechanism has been evaluated using simulations.
pp. 63–71.
We have shown that MRC scales well: 3 or 4
[5] D.Watson, F. Jahanian, and C. Labovitz,
backup configurations is typically enough to isolate
―Experiences with monitoring OSPF on a regional
all links and nodes in our test topologies. MRC
service provider network,‖ in Proc. 23rd Int. Conf.
backup path lengths are comparable to the optimal Distributed Computing Systems (ICDCS’03),
backup path lengths—MRC backup paths are Washington, DC, 2003, pp. 204–213, IEEE
typically zero to two hops longer. Computer Society.
We have evaluated the effect MRC has on [6] P. Francois, C. Filsfils, J. Evans, and O.
the load distribution in the network while traffic is Bonaventure, ―Achieving sub-second IGP
routed in the backup configurations, and we have convergence in large IP networks,‖ ACM

proposed a method that minimizes the risk of SIGCOMM Comput. Commun. Rev., vol. 35, no. 2,
pp. 35–44, Jul. 2005.
congestion after a link failure if we have an estimate
[7] A. Markopoulou, G. Iannaccone, S.
of the demand matrix. In the COST239 network,
Bhattacharyya, C.-N. Chuah, and C. Diot,
this approach gave a maximum link load after the
―Characterization of failures in an IP backbone
worst case link failure that was even lower than
network,‖ in Proc. IEEE INFOCOM, Mar. 2004,
after a full IGP re-convergence on the altered vol. 4, pp. 2307–2317.
topology. MRC thus achieves fast recovery with a [8] S. Nelakuditi, S. Lee, Y. Yu, Z.-L. Zhang, and
very limited performance penalty. C.-N. Chuah, ―Fast local rerouting for handling
7. REFERENCES transient link failures,‖ IEEE/ACM Trans.
[1] D. D. Clark, ―The design philosophy of Networking, vol. 15, no. 2, pp. 359–372, Apr. 2007.
theDARPAinternet protocols,‖ ACM SIGCOMM
Comput. Commun. Rev., vol. 18, no. 4, pp. 106–114,
Aug. 1988.
[2] A. Basu and J. G. Riecke, ―Stability issues in
OSPF routing,‖ in Proc. ACM SIGCOMM, San
Diego, CA, Aug. 2001, pp. 225–236.
[3] C. Labovitz, A. Ahuja, A. Bose, and F. Jahanian,
―Delayed internet routing convergence,‖ IEEE/ACM
Trans. Networking, vol. 9, no. 3, pp. 293–306, Jun.
2001.

Sk.Meeravali1 IJECS Volume 2 Issue 8 August, 2013 Page No.2628-2634 Page 2634